You are on page 1of 588

PRELIMINARY TECHNICAL SUPPORT

DOCUMENT:
ENERGY EFFICIENCY PROGRAM FOR
CONSUMER PRODUCTS AND COMMERCIAL
AND INDUSTRIAL EQUIPMENT:
BATTERY CHARGERS AND EXTERNAL POWER SUPPLIES

September 2010

U.S. Department of Energy


Assistant Secretary
Office of Energy Efficiency and Renewable Energy
Building Technologies Program
Appliances and Commercial Equipment Standards
Washington, DC 20585

This Document was prepared for the Department of Energy

and by staff members of Navigant Consulting, Inc.,

D&R International, Ltd.,

and Ernest Orlando Lawrence Berkeley National Laboratory

TECHNICAL SUPPORT DOCUMENT:


BATTERY CHARGERS AND EXTERNAL POWER SUPPLIES
Table of Contents
CHAPTERS
Executive Summary
Chapter 1

Introduction

Chapter 2

Analytical Framework, Comments from Interested Parties, and


DOE Responses

Chapter 3

Market and Technology Assessment

Chapter 4

Screening Analysis

Chapter 5

Engineering Analysis

Chapter 6

Product Price Determination

Chapter 7

Energy Use Analysis

Chapter 8

Life-Cycle Cost and Payback Period Analyses

Chapter 9

Shipments Analysis

Chapter 10

National Impact Analysis

Chapter 11

Life-Cycle Cost Subgroup Analysis

Chapter 12

Preliminary Manufacturer Impact Analysis

Chapter 13

Utility Impact Analysis

Chapter 14

Employment Impact Analysis

REPORTS
Report

Environmental Assessment for Battery Chargers and External


Power Supplies

Report

Regulatory Impact Analysis for Battery Chargers and External


Power Supplies

APPENDICES
Appendix 3A

Battery Charger and External Power Supply Applications

Appendix 3B

Battery Charger and External Power Supply Efficiency


Programs

Appendix 3C

Evaluation Methods for Differentiating Wall Adapters with


Charge Control

Appendix 5A

External Power Supply Test Data

Appendix 5B

Battery Charger Test Data

Appendix 5C

Manufacturer Interview Guide

Appendix 7A

Battery Charger and External Power Supply Usage Profiles,


Application States, and Loading Points

Appendix 7B

Battery Charger and External Power Supply Application-Level


Unit Energy Consumption Values

Appendix 8A

User Instructions for LCC and PBP Spreadsheets

Appendix 8B

Supplementary LCC and PBP Results

Appendix 8C

End-Use Application Inputs for the LCC

Appendix 8D

Residential Discount Rate Distributions

Appendix 10A

Net Present Value under Alternative Electricity Price Scenarios

Appendix 12A

Preliminary MIA Manufacturer Interview Guide

Appendix 15A

Interagency Report on the Social Cost of Carbon

EXECUTIVE SUMMARY

ES.1
ES.2
ES.3

ES.4

TABLE OF CONTENTS

OVERVIEW OF PRELIMINARY ACTIVITIES.............................................................. 1

OVERVIEW OF THE PRELIMINARY ANALYSES AND THE PRELIMINARY

TECHNICAL SUPPORT DOCUMENT............................................................................ 3

KEY RESULTS OF THE ANALYSIS .............................................................................. 4

ES.3.1 Determination of Scope ....................................................................................... 4

ES.3.2 Market and Technology Assessment ................................................................... 5

ES.3.3 Screening Analysis............................................................................................. 12

ES.3.4 Engineering Analysis ......................................................................................... 12

ES.3.5 Product Price Determination.............................................................................. 18

ES.3.6 Energy Use Analysis.......................................................................................... 21

ES.3.7 Life-Cycle Cost and Payback Period Analyses ................................................. 23

ES.3.8 Shipments Analysis............................................................................................ 31

ES.3.9 National Impact Analysis................................................................................... 31

ES.3.9.1 National Energy Savings Analysis......................................................... 33

ES.3.9.2 Net Present Value Analysis ................................................................... 36

ES.3.10 Preliminary Manufacturer Impact Analysis....................................................... 41

ES.3.11 Other Analyses................................................................................................... 41

ISSUES ON WHICH DOE SEEKS PUBLIC COMMENT............................................. 41

ES.4.1 Criteria for Identifying Charge Control in Wall Adapters................................. 42

ES.4.2 Design Options for Battery Chargers and External Power Supplies ................. 42

ES.4.3 Product Classes for Battery Chargers ................................................................ 42

ES.4.4 Dissimilarities Between Low Energy, High Voltage Battery Chargers ............ 42

ES.4.5 Scaling Within Battery Charger Product Classes .............................................. 42

ES.4.6 Medium Energy, Low Voltage Representative Unit for BCs ............................ 42

ES.4.7 EPS Representative Units .................................................................................. 43

ES.4.8 EPS Max-Tech Value for 2.5-Watt Representative Unit................................... 43

ES.4.9 Efficiency Metrics for Multiple-Voltage EPSs.................................................. 43

ES.4.10 Representative Product Classes and Units for MADB EPSs............................. 43

ES.4.11 Baseline Candidate Standard Level of Efficiency for Medical and MADB EPSs

............................................................................................................................ 44

ES.4.12 Scaling for EPS Representative Units and CSLs ............................................... 44

ES.4.13 Cost Estimates for Product Classes not Fully Analyzed.................................... 45

ES.4.14 Market Data ....................................................................................................... 45

ES.4.15 Product Lifetimes............................................................................................... 45

ES.4.16 Estimated Markups and Incremental Markup Assumptions.............................. 45

ES.4.17 Sensitivity Analysis Methodology and Results ................................................. 45

ES.4.18 Use in the Commercial Sector ........................................................................... 46

ES.4.19 Prospects for Substitution in Response to Standards......................................... 46

ES.4.20 Usage Profiles .................................................................................................... 46

ES.4.21 Selecting Battery Charger Usage Profiles for Use in Determining Compliance

with Standards.................................................................................................... 46

ES.4.22 Shipment Trends ................................................................................................ 47

ES.4.23 Base Case Efficiency Distributions ................................................................... 47

ES.4.24
ES.4.25
ES.4.26
ES.4.27

Quantifying Non-Consumer Benefits ................................................................ 47

Impact on Manufacturers ................................................................................... 47

Cumulative Regulatory Burdens........................................................................ 47

Valuing Airborne Emission Reductions ............................................................ 47

LIST OF TABLES

TABLE ES.3.1 EXTERNAL POWER SUPPLY PRODUCT CLASSES ..................................... 5

TABLE ES.3.2 BATTERY CHARGER PRODUCT CLASSES................................................... 6

TABLE ES.3.3 SUMMARY OF EPS SHIPMENTS ..................................................................... 7

TABLE ES.3.4 CLASS A EXTERNAL POWER SUPPLY SHIPMENTS BY PRODUCT

CLASS AND SEGMENT................................................................................................... 7

TABLE ES.3.5 NON-CLASS A EXTERNAL POWER SUPPLY SHIPMENTS BY PRODUCT

CLASS ................................................................................................................................ 8

TABLE ES.3.6 SUMMARY OF BATTERY CHARGER SHIPMENTS ................................... 10

TABLE ES.3.7 BATTERY CHARGER SHIPMENTS BY PRODUCT CLASS........................ 10

TABLE ES.3.8 COST AND EFFICIENCY RELATIONSHIP FOR 2.5W EPS

(MANUFACTURER INTERVIEWS) ............................................................................. 13

TABLE ES.3.9 COST AND EFFICIENCY RELATIONSHIP FOR 18W EPS

(MANUFACTURER INTERVIEWS) ............................................................................. 13

TABLE ES.3.10 COST AND EFFICIENCY RELATIONSHIP FOR 60W EPS

(MANUFACTURER INTERVIEWS) ............................................................................. 14

TABLE ES.3.11 COST AND EFFICIENCY RELATIONSHIP FOR 120W EPS

(MANUFACTURER INTERVIEWS) ............................................................................. 14

TABLE ES.3.12 COST AND EFFICIENCY RELATIONSHIP FOR BC PRODUCT CLASS 1:

LOW ENERGY, INDUCTIVE ........................................................................................ 15

TABLE ES.3.13 COST AND EFFICIENCY RELATIONSHIP FOR BC PRODUCT CLASS 2:

LOW ENERGY, LOW VOLTAGE ................................................................................. 15

TABLE ES.3.14 COST AND EFFICIENCY RELATIONSHIP FOR BC PRODUCT CLASS 3:

LOW ENERGY, MEDIUM VOLTAGE.......................................................................... 15

TABLE ES.3.15 COST AND EFFICIENCY RELATIONSHIP FOR BC PRODUCT CLASS 4:

LOW ENERGY, HIGH VOLTAGE ................................................................................ 16

TABLE ES.3.16 COST AND EFFICIENCY RELATIONSHIP FOR BC PRODUCT CLASS 5:

MED. ENERGY, LOW VOLTAGE ................................................................................ 16

TABLE ES.3.17 COST AND EFFICIENCY RELATIONSHIP FOR BC PRODUCT CLASS 6:

MED. ENERGY, HIGH VOLTAGE................................................................................ 16

TABLE ES.3.18 COST AND EFFICIENCY RELATIONSHIP FOR BC PRODUCT CLASS 7:

HIGH ENERGY ............................................................................................................... 17

TABLE ES.3.19 COST AND EFFICIENCY RELATIONSHIP FOR BC PRODUCT CLASS 8:

LOW ENERGY, LOW VOLTAGE DC INPUT.............................................................. 17

TABLE ES.3.20 COST AND EFFICIENCY RELATIONSHIP FOR BC PRODUCT CLASS 9:

LOW ENERGY, HIGH VOLTAGE DC INPUT ............................................................. 17

TABLE ES.3.21 COST AND EFFICIENCY RELATIONSHIP FOR BC PRODUCT CLASS 10:

LOW ENERGY, AC OUTPUT........................................................................................ 18

TABLE ES.3.22 EXTERNAL POWER SUPPLY MARKUPS BY PRODUCT CLASS ........... 20

TABLE ES.3.23 BATTERY CHARGER MARKUPS BY PRODUCT CLASS ........................ 21

ii

TABLE ES.3.24 CLASS A EXTERNAL POWER SUPPLY UNIT ENERGY CONSUMPTION


(KWH/YR)........................................................................................................................ 21

TABLE ES.3.25 BATTERY CHARGER USAGE PROFILES* ................................................. 22

TABLE ES.3.26 BATTERY CHARGER UNIT ENERGY CONSUMPTION BASED ON

MANUFACTURER-SUPPLIED DATA (KWH/YR) ..................................................... 22

TABLE ES.3.27 BATTERY CHARGER UNIT ENERGY CONSUMPTION BASED ON TEST

AND TEARDOWN DATA (KWH/YR) .......................................................................... 23

TABLE ES.3.28 LCC AND PBP RESULTS FOR 2.5W REGULAR AC/DC EXTERNAL

POWER SUPPLIES (MANUFACTURER DATA)......................................................... 24

TABLE ES.3.29 LCC AND PBP RESULTS FOR 18W REGULAR AC/DC EXTERNAL

POWER SUPPLIES (MANUFACTURER DATA)......................................................... 25

TABLE ES.3.30 LCC AND PBP RESULTS FOR 60W REGULAR AC/DC EXTERNAL

POWER SUPPLIES (MANUFACTURER DATA)......................................................... 25

TABLE ES.3.31 LCC AND PBP RESULTS FOR 120W REGULAR AC/DC EXTERNAL

POWER SUPPLIES (MANUFACTURER DATA)......................................................... 25

TABLE ES.3.32 LCC AND PBP RESULTS FOR PRODUCT CLASS 1: LOW ENERGY,

INDUCTIVE BATTERY CHARGERS (MANUFACTURER DATA) .......................... 26

TABLE ES.3.33 LCC AND PBP RESULTS FOR PRODUCT CLASS 2: LOW ENERGY, LOW

VOLTAGE BATTERY CHARGERS (MANUFACTURER DATA) ............................. 26

TABLE ES.3.34 LCC AND PBP RESULTS FOR PRODUCT CLASS 2: LOW ENERGY, LOW

VOLTAGE BATTERY CHARGERS (TEST/TEARDOWN DATA)............................. 26

TABLE ES.3.35 LCC AND PBP RESULTS FOR PRODUCT CLASS 4: LOW ENERGY,

HIGH VOLTAGE BATTERY CHARGERS (MANUFACTURER DATA) .................. 26

TABLE ES.3.36 LCC AND PBP RESULTS FOR PRODUCT CLASS 4: LOW ENERGY,

HIGH VOLTAGE BATTERY CHARGERS (TEST/TEARDOWN).............................. 27

TABLE ES.3.37 LCC AND PBP RESULTS FOR PRODUCT CLASS 5: MEDIUM ENERGY,

LOW VOLTAGE BATTERY CHARGERS (MANUFACTURER DATA)................... 27

TABLE ES.3.38 LCC AND PBP RESULTS FOR PRODUCT CLASS 5: MEDIUM ENERGY,

LOW VOLTAGE BATTERY CHARGERS (TEST/TEARDOWN DATA) .................. 27

TABLE ES.3.39 LCC AND PBP RESULTS FOR PRODUCT CLASS 7: HIGH ENERGY

BATTERY CHARGERS (MANUFACTURER DATA)................................................. 27

TABLE ES.3.40 LCC AND PBP RESULTS FOR PRODUCT CLASS 10: LOW ENERGY, AC

OUT BATTERY CHARGERS (MANUFACTURER DATA)........................................ 28

TABLE ES.3.41 LCC AND PBP RESULTS FOR PRODUCT CLASS 10: LOW ENERGY, AC

OUT BATTERY CHARGERS (TEST/TEARDOWN DATA) ....................................... 28

TABLE ES.3.42 EPS WEIGHTED-AVERAGE LCC SAVINGS SENSITIVITY ANALYSIS

SUMMARY...................................................................................................................... 29

TABLE ES.3.43 EPS WEIGHTED-AVERAGE LCC SAVINGS BY CSL................................ 29

TABLE ES.3.44 BC WEIGHTED-AVERAGE LCC SAVINGS SENSITIVITY ANALYSIS

SUMMARY...................................................................................................................... 30

TABLE ES.3.45 BC WEIGHTED-AVERAGE LCC SAVINGS BY PRODUCT CLASS ........ 31

TABLE ES.3.46 ESTIMATED NATIONAL ANNUAL ENERGY CONSUMPTION OF

EXTERNAL POWER SUPPLIES SHIPPED IN 2008 .................................................... 32

TABLE ES.3.47 ESTIMATED NATIONAL ANNUAL ENERGY CONSUMPTION OF

BATTERY CHARGERS SHIPPED IN 2008 .................................................................. 33

iii

TABLE ES.3.48 NATIONAL ENERGY SAVINGS POTENTIAL FROM AMENDED

STANDARDS FOR CLASS A EXTERNAL POWER SUPPLIES (TRILLION BTU) . 34

TABLE ES.3.49 NATIONAL ENERGY SAVINGS POTENTIAL FROM NEW STANDARDS

FOR BATTERY CHARGERS (TRILLION BTU), MANUFACTURER DATA........... 35

TABLE ES.3.50 NATIONAL ENERGY SAVINGS POTENTIAL FROM NEW STANDARDS

FOR BATTERY CHARGERS (TRILLION BTU), TEST AND TEARDOWN DATA. 36

TABLE ES.3.51 NET PRESENT VALUE OF AMENDED STANDARDS FOR CLASS A

EXTERNAL POWER SUPPLIES AT A 3 PERCENT DISCOUNT RATE (2009$

MILLIONS) ...................................................................................................................... 37

TABLE ES.3.52 NET PRESENT VALUE OF AMENDED STANDARDS FOR EXTERNAL

POWER SUPPLIES AT A 7 PERCENT DISCOUNT RATE (2009$ MILLIONS) ....... 37

TABLE ES.3.53 NET PRESENT VALUE OF NEW STANDARDS FOR BATTERY

CHARGERS AT A 3 PERCENT DISCOUNT RATE (2009$ MILLIONS),

MANUFACTURER DATA ............................................................................................. 38

TABLE ES.3.54 NET PRESENT VALUE OF NEW STANDARDS FOR BATTERY

CHARGERS AT A 7 PERCENT DISCOUNT RATE (2009$ MILLIONS),

MANUFACTURER DATA ............................................................................................. 39

TABLE ES.3.55 NET PRESENT VALUE OF NEW STANDARDS FOR BATTERY

CHARGERS AT A 3 PERCENT DISCOUNT RATE (2009$ MILLIONS), TEST AND

TEARDOWN DATA........................................................................................................ 40

TABLE ES.3.56 NET PRESENT VALUE OF NEW STANDARDS FOR BATTERY

CHARGERS AT A 7 PERCENT DISCOUNT RATE (2009$ MILLIONS), TEST AND

TEARDOWN DATA........................................................................................................ 40

LIST OF FIGURES

FIGURE ES.1.1 FLOW DIAGRAM OF ANALYSES FOR THE BATTERY CHARGER AND

EXTERNAL POWER SUPPLY RULEMAKING PROCESS .......................................... 2

FIGURE ES.3.1 DISTRIBUTION CHANNELS FOR BATTERY CHARGERS AND

EXTERNAL POWER SUPPLIES ................................................................................... 19

iv

EXECUTIVE SUMMARY

ES.1

OVERVIEW OF PRELIMINARY ACTIVITIES

By statute, the U.S. Department of Energy (DOE) must establish energy conservation
standards that are technologically feasible and economically justified, and would achieve the
maximum improvement in energy efficiency. 42 U.S.C. 6295(o)(2)(B). This executive
summary describes the preliminary activities that DOE conducted in consideration of energy
conservation standards for battery chargers (BCs) and external power supplies (EPSs) and
summarizes key results from DOEs preliminary analyses. Additionally, the executive summary
identifies specific issues about which DOE seeks comments from interested parties. These issues
are listed in section 4 of the executive summary, which are highlighted in the public meeting
presentation, and discussed further in the relevant sections of the preliminary technical support
document (TSD).
Figure ES.1.1 presents a summary of the analytical components of the standards-setting
process and illustrates how key results are generated. The focal point of the figure is the center
column, labeled Analyses. The columns labeled Key Inputs and Key Outputs show how
the analyses fit into the process and how they relate to each other. Key inputs are the types of
data and other information that the analyses require. Some key information is obtained from
public sources but others may be collected from interested parties or persons having special
knowledge and expertise. Key outputs are analytical results that feed directly into the standards
setting process. The issues on which DOE seeks comment from interested parties derive from the
key results that are generated by the preliminary analyses. Arrows connecting analyses show the
types of information that feed from one analysis into another.

Approaches

Key Inputs

Key Outputs

Analyses

PublishFramework
FrameworkDocument
Document
Characterize Industry

Identify Firms/Products
Historical Shipments

Analysis of Market Data

Market Segmentation
Non-Regulatory Programs

Analysis of Product Data

Product Prototypes

Efficiency-Level Approach
Design Option Approach

Commercial Products

Technology Options

Product Classes

Energy Use

Design Options

Screening Analysis
Design Options

Efficiency/Performance
Analysis of Energy Use
Data

Product Classes
Technology Options

Market and Technology


Assessment

Product Designs

Engineering Analysis
Design Energy/ Use
Energy-Efficiency
Levels

Product Designs
Annual Energy Use (UEC)

Retail Price Collection and


Analysis

Product Price
Determination

Retail
Prices

Energy-Efficiency Levels
Accounting Approach
Backcast and Forecast
Market Saturation

Shipments
Analysis

Life-Cycle Costs
Payback Periods

Life-Cycle Cost and


Payback Period
Analyses

Energy Prices
Installation Costs
Maintenance & Repair Costs

Candidate
Standard
Levels

UEC
Energy Prices

Installation
Costs
Maint Costs
Repair Costs
National Energy Savings
Net Present Values

National Impact
Analysis

Energy Price Forecasts


Site-to-Source Factors

Conversion Capital Expenditures


Direct Employment Impacts

Preliminary
Manufacturer Impact
Analysis

Manufacturer Prices
Average Costs

Advance Notice of Proposed


RulemakingAnalyses
(ANO)
Publish Preliminary

Revise Preliminary
Analyses

Stakeholder Comments

Demographics

Manufacturer Interviews

Manufacturer Prices
Average Costs

GRIM Analysis

Manufacturer Financial Data

NEMS-BT

Utility Load Factors


National Energy Savings

ImSET

National Energy Savings


National Product Costs
National Operating Costs

NEMS-BT

Life-Cycle Cost
Sub-Group Analysis
Manufacturer Impact
Analysis
Utility Impact
Analysis
Employment Impact
Analysis

Emission Rates
National Energy Savings

Environmental
Assessment

Non-Regulatory
Alternatives

Regulatory Impact
Analysis

Trial Standard Levels (TSLs)


TSLs
Life-Cycle Costs
Payback Periods
Industry Cash Flow
Sub-Group Cash-Flow
Direct Employment Impacts
Competitive Impacts
Cumulative Regulatory Burden
Direct Employment
Impacts
Utility Impacts

National Employment Impacts

Emission Estimates
National Energy Savings
Net Present Values

Publish Notice of Proposed Rulemaking (NOPR)


(NO))
Department of Justice Review

Revise Analyses

Stakeholder Comments

Revised Results

Final Rule

Figure ES.1.1 Flow Diagram of Analyses for the Battery Charger and External Power
Supply Rulemaking Process

ES.2 OVERVIEW OF THE PRELIMINARY ANALYSES AND THE PRELIMINARY


TECHNICAL SUPPORT DOCUMENT
For the preliminary stage, DOE publishes a notice of public meeting (NOPM) in the
Federal Register, which announces the availability of the preliminary technical support
document, the date and place of the public meeting, and presentation materials interested parties
may review before the public meeting. In addition, the NOPM summarizes the major analyses
DOE developed in the preliminary stage of the rulemaking. The preliminary TSD describes each
preliminary analysis in detail, providing detailed descriptions of inputs, sources, methodologies,
and results.
Chapter 2 provides an overview of each preliminary analysis, the comments received on
the analytical approaches DOE described in the framework document, and DOEs responses to
those comments. The following chapters of the preliminary TSD address the analyses performed
for the preliminary stage of the rulemaking.
A market and technology assessment characterizes the relevant product markets and
technology options, including prototype designs (chapter 3).
A screening analysis reviews each technology option to determine whether it is
technologically feasible; is practicable to manufacture, install, and service; would adversely
affect product utility or product availability; or would have adverse impacts on health and safety
(chapter 4).
An engineering analysis develops cost-efficiency relationships that show a
manufacturers cost of achieving increased efficiency. DOE uses manufacturer markups to
convert bill of materials (BOM) cost and manufacturer production cost (MPC) to manufacturer
selling price (MSP) (chapter 5).
A markups analysis converts the manufacturer costs derived in the engineering analysis
to final product prices (chapter 6).
An energy use analysis determines the annual energy use in the field of the considered
products (chapter 7).
Life-cycle cost (LCC) and payback period (PBP) analyses calculate, at the consumer
level, the discounted savings in operating costs throughout the estimated average life of the
covered products, compared to any increase in the products installed cost likely to result directly
from the imposition of a given standard. DOE calculated three sets of LCC results to determine
potential savings based on a reference case and to characterize possible variations in a highsavings case and a low-savings case (chapter 8).
A shipments analysis forecasts product shipments. That forecast is then used to calculate
the impacts of standards on national energy consumption, the net present value (NPV) of
consumer costs and savings, and future manufacturer cash flows (chapter 9).

A national impact analysis (NIA) assesses the cumulative national energy savings (NES)
from standards and the NPV of consumer costs and savings associated with standards at different
efficiency levels (chapter 10).
A preliminary manufacturer impact analysis (MIA) assesses the potential impacts of
energy conservation standards on manufacturers, such as effects on expenditures for capital
conversion, marketing costs, shipments, and research and development costs (chapter 12).
The remaining chapters of the preliminary TSD address the analyses to be performed at
the NOPR stage:
An LCC consumer subgroup analysis evaluates the effects of energy conservation standards
on various national subgroups of the population (chapter 11).
A utility impact analysis examines impacts of energy conservation standards on the
generation capacity of electric utilities (chapter 13).
An employment impact analysis examines the effects of energy conservation standards on
national employment (chapter 14).
An environmental assessment examines the effects of energy conservation standards on
various airborne emissions (chapter 15).
A regulatory impact analysis examines the national impacts of non-regulatory alternatives to
mandatory energy conservation standards (chapter 16).
ES.3

KEY RESULTS OF THE ANALYSIS


The following sections describe in detail the key analyses DOE has performed to date.

ES.3.1 Determination of Scope


This standards rulemaking concerns all BCs and EPSs, which are defined as follows:
The term battery charger means a device that charges batteries for consumer
products, including battery chargers embedded in other consumer products. (42
U.S.C. 6291(32))
The term external power supply means an external power supply circuit that is used
to convert household electric current into DC current or lower-voltage AC current to
operate a consumer product. (42 U.S.C. 6291(36)(A))
Non-Class A EPSs were not initially within the scope of this standards rulemaking.
However, DOE recently determined that standards are warranted for these products in a final
determination published on May 14, 2010. 75 FR 27170. As a result, non-Class A EPSs are
included in the preliminary market and technology assessment, markup analysis, and shipments
analysis. The preliminary engineering analysis for non-Class A EPSs includes representative
4

product classes, representative units, and CSLs, but does not include non-Class A EPS costefficiency outputs. Hence, non-Class A EPSs are not included in those analyses that make use of
engineering outputs, i.e., the LCC and NIA. DOE intends to conduct a comprehensive analysis of
potential standards for these products in the NOPR stage of the rulemaking. See chapter 2 for
further discussion of scope.
ES.3.2 Market and Technology Assessment
A market and technology assessment characterizes the relevant product markets and
technology options, including prototype designs. See chapter 3 for details.
DOE generally divides covered products into product classes by the type of energy used
or by capacity or other performance-related features that affect efficiency. Different energy
conservation standards may apply to different product classes. (42 U.S.C. 6295(q))
DOE is considering amending standards for Class A EPSs and creating new standards for
non-Class A EPSs. DOE is considering classifying EPSs by output type, output voltage, output
power, and (in the case of certain non-Class A EPSs) application, as shown in Table ES.3.1.
Table ES.3.1 External Power Supply Product Classes
Product Class ID
Class A
DC Output
AC Output

Basic Voltage

A1

Low Voltage

A2

Basic Voltage

A3

Low Voltage

A4

Non-Class A
Multiple-Voltage
High Power
DC Output
For Medical Devices
AC Output
For Motor-Operated
BC Applications and
Detachable Batteries
(MADB)

DC Output
AC Output

100 W

X1

100 W

X2

>250 W

H1

Basic Voltage

M1

Low Voltage

M2

Basic Voltage

M3

Low Voltage

M4

Basic Voltage

B1

Low Voltage

B2

Basic Voltage

B3

Low Voltage

B4

DOE is also considering new standards for BCs. DOE is considering classifying BCs by
input and output type, battery energy, battery voltage, and input voltage, as shown in Table
ES.3.2.
Table ES.3.2 Battery Charger Product Classes
Input-Output

Battery Energy

Battery Voltage

<100 Wh
AC-DC
1003000 Wh

Inductive Connection

<4 V

410 V

>10 V

<20 V

20 V

>3000 Wh
DC-DC
AC-AC

Product Class
ID

7
<9 V Input

9 V Input

AC Output from Battery

10

The assessment addresses (1) manufacturer market share and characteristics, (2) existing
regulatory and non-regulatory initiatives for improving product efficiency, and (3) trends in
product characteristics and retail markets. This information yielded supporting data and resource
material on which DOE relied throughout the analysis.
DOE reviews published literature and interviewed manufacturers to develop an overall
understanding of the BC and EPS industries. Industry publications, including trade journals,
literature from manufacturers, government agencies, and trade organizations provided the bulk of
the information, including: (1) manufacturers and their market shares, (2) shipments by
application, (3) product characteristics, and (4) industry trends. DOE estimates that just over 300
million EPSs and just over 400 million BCs shipped in 2008. A summary of EPS shipments is
presented in Table ES.3.3. A detailed list of shipments by product class for Class A EPSs and
non-Class A EPSs is displayed in Table ES.3.4 and Table ES.3.5, respectively. A summary of
BC shipments is presented in Table ES.3.6. A detailed list of shipments by product class for BCs
is displayed in Table ES.3.7.

Table ES.3.3 Summary of EPS Shipments


2008 EPS Shipments
(thousands)
206,176
72,195
7,994
2,250
12,405
301,021

EPS Product Class


Class A1 (DC Output, Basic Voltage)
Class A2 (DC Output, Low Voltage)
Class A3 (AC Output, Basic Voltage)
Class A4 (AC Output, Low Voltage)
Non-Class A
All EPSs

Table ES.3.4 Class A External Power Supply Shipments by Product Class and Segment
Top Applications by Shipments

2008 EPS
Shipments
(thousands)

Percent of
Shipments

Product Class A1 (DC Output, Basic Voltage): 0-10.25 W


1

Answering Machines

20,175

26.5%

Cordless Phones

19,151

25.1%

Mobile Phones

13,140

17.2%

Portable Video Game Systems

10,884

14.3%

In-Vehicle GPS

7,660

10.0%

Other

5,217

6.8%

Total

76,227

100.0%

Product Class A1 (DC Output, Basic Voltage): 10.25-39 W


1

LAN Equipment

27,581

41.3%

VoIP Adapters

8,845

13.3%

Digital Picture Frames

7,472

11.2%

Portable DVD Players

7,140

10.7%

MP3 Speaker Docks

7,012

10.5%

Other

8,671

13.0%

Total

66,721

100.0%

Product Class A1 (DC Output, Basic Voltage): 39-90 W


1

Notebooks

30,225

57.0%

Video Game Consoles

13,512

25.5%

Ink Jet Imaging Equipment

5,557

10.5%

Netbooks

3,700

7.0%

56,776

100.0%

10,075

98.4%

160

1.6%

10,235

100.0%

Total
Product Class A1 (DC Output, Basic Voltage): 90-250 W
1

Notebooks

LED Monitors
Total

Top Applications by Shipments

2008 EPS
Shipments
(thousands)

Percent of
Shipments

Product Class A2 (DC Output, Low Voltage)


1

Mobile Phones

52,560

72.8%

In-Vehicle GPS

7,660

10.6%

MP3 Players

4,373

6.1%

Portable Video Game Systems

2,893

4.0%

Personal Digital Assistants

1,582

2.2%

Other

3,127

4.3%

Total

72,195

100.0%

Product Class A3 (AC Output, Basic Voltage)


1

Home Security Systems

4,219

52.8%

Aquarium Accessories

1,750

21.9%

Water Softeners/Purifiers

1,150

14.4%

Indoor Fountains

500

6.3%

Irrigation Timers

375

4.7%

7,994

100.0%

1,750

77.8%

500

22.2%

2,250

100.0%

Total
Product Class A4 (AC Output, Low Voltage)
1

Aquarium Accessories

Indoor Fountains
Total

Table ES.3.5 Non-Class A External Power Supply Shipments by Product Class


2008 EPS
Percent of
Top Applications by Shipments
Shipments
Shipments
(thousands)
Product Class X1(Multiple-Voltage)
1

Ink Jet Imaging Equipment

3,782

100.0%

Total

3,782

100.0%

Gaming System (multi-voltage EPS)

4,901

100.0%

Total

4,901

100.0%

Amateur Radios

100.0%

Total

100.0%

1,000

65.2%

450

29.3%

50

3.3%

Product Class X2 (Multiple Voltage)


1

Product Class H1(High Power)


1

Product Class M1(Medical Devices)


1

Sleep Apnea Machines

Medical Nebulizers

Portable O2 Concentrators - Others

Top Applications by Shipments


4

Blood Pressure Monitors

Portable O2 Concentrators - Higher Output

2008 EPS
Percent of
Shipments
Shipments
(thousands)

Total

25

1.6%

0.6%

1,534

Product Class M2 (Medical Devices)


1

Blood Pressure Monitors

25

100.0%

Total

25

100.0%

Product Class M3 (Medical Devices)


[no products identified]
Total

Product Class M4 (Medical Devices)


[no products identified]
Total

Product Class B1(MADB)


1

Toy Ride-On Vehicles

355

18.8%

RC Toys

350

18.6%

DIY Power Tools (External)

281

14.9%

DIY Power Tools (Integral)

234

12.4%

Handheld Vacuums

209

11.1%

Other

457

24.2%

Total

1,884

100.0%

164

59.7%

Product Class B2 (MADB)


1

Shavers

Beard and Mustache Trimmers

59

21.3%

Hair Clippers

38

13.8%

Can Openers
Total

14

5.2%

275

100.0%

Product Class B3 (MADB)


[no products identified]
Total

Product Class B4 (MADB)


[no products identified]
Total

Table ES.3.6 Summary of Battery Charger Shipments


BC Product Class
Class 1 (Low Energy, Inductive)
Class 2 (Low Energy, Low Voltage)
Class 3 (Low Energy, Medium Voltage)
Class 4 (Low Energy, High Voltage)
Class 5 (Medium Energy, Low Voltage)
Class 6 (Medium Energy, High Voltage)
Class 7 (High Energy)
Class 8 (Low Energy, <9 V Input)
Class 9 (Low Energy, 9 V Input)
Class 10 (Low Energy, AC Output from Battery)
Total

2008 BC Shipments
(thousands)
5,354
206,221
23,116
69,758
2,333
940
214
71,825
15,519
6,900
402,180

Table ES.3.7 Battery Charger Shipments by Product Class


2008 BC
Percent of
Top Applications by Shipments
Shipments
Shipments
(thousands)
Product Class 1 (Low Energy, Inductive)
1

Rechargeable Toothbrushes

Rechargeable water Jets


Total

4,868

90.9%

487

9.1%

5,354

100.0%

105,120

51.0%

Product Class 2 (Low Energy, Low Voltage)


1

Mobile Phones

Answering Machines

20,175

9.8%

Cordless Phones

19,151

9.3%

Portable Video Game Systems

13,777

6.7%

Digital Cameras

10,879

5.3%

Other

37,119

18.0%

Total

206,221

100.0%

Product Class 3 (Low Energy, Medium Voltage)


1

Portable DVD Players

7,140

30.9%

Camcorders

4,206

18.2%

Toy Ride-On Vehicles

3,548

15.3%

RC Toys

2,100

9.1%

DIY Power Tools (External)

1,753

7.6%

Other

4,369

18.9%

Total

23,116

100.0%

10

Top Applications by Shipments

2008 BC
Percent of
Shipments
Shipments
(thousands)

Product Class 4 (Low Energy, High Voltage)


1

Notebooks

40,300

57.8%

Professional Power Tools

11,688

16.8%

DIY Power Tools (External)

5,259

7.5%

Netbooks

3,700

5.3%

Handheld Vacuums

2,797

4.0%

Other

6,014

8.6%

Total

69,758

100.0%

1,774

76.0%

Product Class 5 (Medium Energy, Low Voltage)


1

Toy Ride-On Vehicles

Marine/Automotive/RV Chargers

500

21.4%

Portable O2 concentrator - Others

50

2.1%

Portable O2 concentrator - Higher output

0.4%

2,333

100.0%

Total
Product Class 6 (Medium Energy, High Voltage)
1

Electric Scooters

250

26.6%

Mobility Scooters

192

20.4%

Lawn Mowers

182

19.4%

Wheelchairs

166

17.7%

Motorized Bicycles

150

15.9%

Total

940

100.0%

Golf Carts

214

100.0%

Total

214

100.0%

Product Class 7 (High Energy)


1

Product Class 8 (Low Energy, <9 V Input)


1

MP3 Players

39,358

54.8%

Mobile Phones

26,280

36.6%

Digital Cameras

2,720

3.8%

Personal Digital Assistants

1,779

2.5%

Camcorders

1,402

2.0%

Other

286

0.4%

Total

71,825

100.0%

15,320

98.7%

Product Class 9 (Low Energy, 9 V Input)


1

In-Vehicle GPS

Medical Nebulizer

90

0.6%

Portable O2 concentrator Others

50

0.3%

Flashlights/Lanterns

50

0.3%

11

Top Applications by Shipments


5

2008 BC
Percent of
Shipments
Shipments
(thousands)

Portable O2 concentrator - Higher


output
Total

0.1%

15,519

100.0%

Uninterruptible Power Supplies

6,900

100.0%

Total

6,900

100.0%

Product Class 10 (Low Energy, AC Output from Battery)


1

DOE typically uses information about existing and past technology options and prototype
designs to determine which technologies and combinations of technologies manufacturers use to
attain higher performance levels. In consultation with interested parties, DOE develops a list of
technologies to be considered. The identified technologies initially include all those DOE
believes are technologically feasible.
ES.3.3 Screening Analysis
DOE also develops an initial list of options for enhancing efficiency from the
technologies that the technology assessment identified as feasible. In consultation with interested
parties, DOE reviews the list to assess whether the technologies are practicable to manufacture,
install, and service; would adversely affect product utility or availability; or would have adverse
impacts on health and safety. In the subsequent engineering analysis, DOE further examines the
technology options that it did not remove from consideration in the screening analysis. See
chapter 4 of the preliminary TSD for details.
ES.3.4 Engineering Analysis
The engineering analysis develops cost-efficiency relationships that show a
manufacturers cost of achieving increased efficiency. DOE uses manufacturer markups to
convert manufacturer production cost (MPC) to manufacturer selling price (MSP) (chapter 5).
DOE structured its engineering analysis around two methodologies: (1) test and teardown
method, which involves testing products for efficiency and determining cost from a detailed bill
of materials derived from tear-downs and (2) the efficiency-level approach, whereby
manufacturers provide and explain their costs of achieving increases in energy efficiency at
discrete levels of efficiency.
For the Class A external power supplies engineering analysis, DOE analyzed the four
representative units in product class A1. For each representative unit, DOE chose a baseline level
(CSL 0), and identified four more efficient replacements: ENERGY STAR 2.0, intermediate,
best-in-market, and max-tech (CSL 1, CSL 2, CSL 3, CSL 4). For each CSL, DOE calculated the
manufacturer selling price, measured the average active-mode efficiency and the no-load power,
and used the values to generate cost-efficiency curves for each representative unit. DOE only
calculated CSL 4 values from the manufacturer interview method and not the test and teardown
method, as a max-tech unit is inherently not commercially available.
12

DOE developed two sets of cost-efficiency curves for EPSs: one based on manufacturer data and
another based on test and teardown data. In summary, for each representative unit, the
manufacturer data for cost versus efficiency and cost versus no-load power data showed a trend
of increasing MSPs with higher CSLs, meaning that it costs more to manufacture more efficient
EPSs. On the other hand, the test and teardown curves cost-efficiency relationship was
inconclusive. In many cases for the test and teardown data, the CSL 3 best-in-market unit was
the cheapest, and the CSL 0 unit was the most expensive. For all four test and teardown curves,
representative units cost decreased from CSL 1 to CSL 3.
DOE believes that the results from the test and teardown method are inconclusive
because they do not show a clear relationship. One reason for this may be that DOE did not
sufficiently control for all factors affecting cost and efficiency, as explained further in chapter 5.
Therefore, DOE chose to use only the manufacturer data in its UEC, LCC, and NIA analysis for
Class A EPSs. For the three representative units test and teardown curves where CSL 3 is the
least expensive, the LCC savings are inherently positive at CSL 3. Table ES.3.8, Table ES.3.9,
Table ES.3.10, and Table ES.3.11 present the efficiency, no-load power, and MSP difference
identified for each CSL in the engineering analysis for manufacturer data. The tables present
both manufacturer data and CSL data in parentheses. All manufacturer data points qualify for the
CSLs by either meeting or exceeding the CSL efficiency requirements. The MSP difference is
the incremental price of the EPS above the baseline model.
Table ES.3.8 Cost and Efficiency Relationship for 2.5W EPS (Manufacturer Interviews)
CSL 0

CSL 1

CSL 2

CSL 3

CSL 4

Mfr Data Efficiency [%]:

58.3%

67.9%

71.0%

73.5%

74.0%

Mfr Data No Load Power [W]:

0.500

0.300

0.130

0.100

0.053

CSL Description:

EISA

E. STAR 2.0

Intermediate

Best in Market

Max Tech

CSL Eff. [%],

(58.3%)

(67.9%)

(70.9%)

(73.2%)

(73.9%)

No-Load Power [W]

(0.500)

(0.300)

(0.200)

(0.100)

(0.062)

MSP Difference[$]:

$0.00

$0.04

$0.23

$0.31

$0.42

Table ES.3.9 Cost and Efficiency Relationship for 18W EPS (Manufacturer Interviews)
CSL 0

CSL 1

CSL 2

CSL 3

CSL 4

Mfr Data Efficiency [%]:

76.0%

80.3%

83.0%

85.4%

87.5%

Mfr Data No Load Power [W]:

0.500

0.300

0.200

0.100

0.062

CSL Description:

EISA

E. STAR 2.0

Intermediate

Best in Market

Max Tech

CSL Eff. [%],

(76.0%)

(80.3%)

(83.0%)

(85.0%)

(87.3%)

No-Load Power [W]

(0.500)

(0.300)

(0.200)

(0.100)

(0.062)

MSP Difference[$]:

$0.00

$0.32

$0.42

$0.79

$1.23

13

Table ES.3.10 Cost and Efficiency Relationship for 60W EPS (Manufacturer Interviews)
CSL 0

CSL 1

CSL 2

CSL 3

CSL 4

Mfr Data Efficiency [%]:

85.0%

87.0%

87.0%

88.0%

91.0%

Mfr Data No Load Power [W]:

0.500

0.500

0.200

0.073

0.073

CSL Description:

EISA

E. STAR 2.0

Intermediate

Best in Market

Max Tech

CSL Eff. [%],

(85.0%)

(87.0%)

(87.0%)

(88.0%)

(91.0%)

No-Load Power [W]

(0.500)

(0.500)

(0.230)

(0.210)

(0.165)

MSP Difference[$]:

$0.00

$0.24

$0.71

$1.58

$2.90

Table ES.3.11 Cost and Efficiency Relationship for 120W EPS (Manufacturer Interviews)
CSL 0

CSL 1

CSL 2

CSL 3

CSL 4

Mfr Data Efficiency [%]:

85.0%

87.0%

88.0%

88.4%

91.7%

Mfr Data No Load Power [W]:


CSL Description:

0.500
EISA

0.500
E. STAR 2.0

0.230
Intermediate

0.210
Best in Market

0.165
Max Tech

CSL Eff. [%],

(85.0%)

(87.0%)

(87.0%)

(88.0%)

(91.0%)

No-Load Power [W]

(0.500)

(0.500)

(0.230)

(0.210)

(0.165)

MSP Difference[$]:

$0.00

$0.66

$1.23

$1.41

$5.03

For the battery charger engineering analysis, DOE analyzed 10 product classes that cover
the entire range of consumer BCs available in the market. For each product class, DOE chose
baseline battery chargers (CSL 0) representative of the least efficient in the market and identified
more efficient replacements: improved, best-in-market, and max-tech (CSL 1, CSL 2, CSL 3).
For each CSL, DOE calculated the manufacturer selling price and shipment-weighted average
annual energy consumption, and used the values to generate cost-efficiency curves. DOE only
calculated CSL 3 values from the manufacturer interview method and not the test and teardown
method, as a max-tech unit is inherently not commercially available.
DOE collected data for 6 of the 10 BC product classes. For the four product classes for
which DOE did not collect data directly, DOE developed a scaling method to estimate MSP and
energy consumption values from the results for the other product classes. See chapter 5 of the
preliminary TSD for details.
DOE developed cost-efficiency curves for BCs based on test and teardown data as well as
manufacturer data. For each representative product class, DOE used either test and teardown
data, manufacturer data, or both to characterize the cost-efficiency relationship for the
representative unit. Table ES.3.12, Table ES.3.13, Table ES.3.14, Table ES.3.15, Table ES.3.16,
Table ES.3.17, Table ES.3.18, Table ES.3.19, Table ES.3.20, and Table ES.3.21 present the
energy consumption and MSP identified for each CSL in the engineering analysis.

14

Table ES.3.12 Cost and Efficiency Relationship for BC Product Class 1: Low Energy,
Inductive
MSP from
MSP from
UEC
Interviews Teardowns
CSL

kWh/yr

2009$

2009$

10.0

$2.05

7.0

$2.22

3.5

$2.45

1.5

$2.60

Table ES.3.13 Cost and Efficiency Relationship for BC Product Class 2: Low Energy, Low
Voltage
MSP from
MSP from
UEC
Interviews Teardowns
CSL

kWh/yr

2009$

2009$

10.0

$0.62

6.0

$1.09

$0.62

1.2

$11.71

$2.62

Table ES.3.14 Cost and Efficiency Relationship for BC Product Class 3: Low Energy,
Medium Voltage
MSP from
MSP from
UEC
Interviews Teardowns
CSL

kWh/yr

2009$

2009$

10.0

$3.16

5.4

$4.96

1.0

$12.99

15

Table ES.3.15 Cost and Efficiency Relationship for BC Product Class 4: Low Energy, High
Voltage
MSP from
MSP from
UEC
Interviews Teardowns
CSL

kWh/yr

2009$

2009$

39.4

$3.79

$3.79

10.5

$9.52

$6.76

6.1

$12.68

$7.44

Table ES.3.16 Cost and Efficiency Relationship for BC Product Class 5: Med. Energy, Low
Voltage
MSP from
MSP from
UEC
Interviews Teardowns
CSL

kWh/yr

2009$

2009$

202.7

$18.48

$18.48

159.6

$27.46

$21.71

100.0

$64.14

75.0

$127.00

Table ES.3.17 Cost and Efficiency Relationship for BC Product Class 6: Med. Energy,
High Voltage
MSP from
MSP from
UEC
Interviews Teardowns
CSL

kWh/yr

2009$

2009$

69.3

$18.48

41.9

$35.87

25.0

$76.83

18.0

$139.95

16

Table ES.3.18 Cost and Efficiency Relationship for BC Product Class 7: High Energy
MSP from
MSP from
UEC
Interviews Teardowns
CSL

kWh/yr

2009$

2009$

290.0

$79.01

250.0

$94.94

200.0

$127.05

150.0

$192.32

Table ES.3.19 Cost and Efficiency Relationship for BC Product Class 8: Low Energy, Low
Voltage DC Input
MSP from
MSP from
UEC
Interviews Teardowns
CSL

kWh/yr

2009$

2009$

1.5

$0.62

0.9

$1.42

0.8

$2.17

Table ES.3.20 Cost and Efficiency Relationship for BC Product Class 9: Low Energy, High
Voltage DC Input
MSP from
MSP from
UEC
Interviews Teardowns
CSL

kWh/yr

2009$

2009$

1.3

$0.62

0.8

$1.11

0.5

$2.76

17

Table ES.3.21 Cost and Efficiency Relationship for BC Product Class 10: Low Energy, AC
Output
MSP from
MSP from
UEC
Interviews Teardowns
CSL

kWh/yr

2009$

2009$

19.6

$2.76

$2.76

6.4

$3.93

$2.11

4.0

$4.25

1.5

$4.64

ES.3.5 Product Price Determination


The product price determination, documented in chapter 6, develops markups at the enduse product manufacturer (referred to as the OEM) and retail stages of the distribution chain to
convert the estimates of the BC/EPS manufacturer selling prices (derived in the engineering
analysis) to consumer prices. In developing markups, DOE determined the distribution channels
for BCs and EPSs and the markup associated with each party in the distribution channel.
Figure ES.3.1 displays the distribution channels for BCs and EPSs. The gray arrows show the
most common path to market.

18

Figure ES.3.1 Distribution Channels for Battery Chargers and External Power Supplies
Because the majority of these products pass through an OEM, where they are integrated
into or packaged with an end-use consumer product, and are later sold through retailers, DOE
assumes that the markups applied at these stages set the final product price. Composite baseline
and incremental markups for each EPS and BC product class are shown in Table ES.3.22 and
Table ES.3.23, respectively. To obtain the final product price, the MSP is multiplied by the
appropriate composite markups and sales tax rate. DOE used a national average sales tax rate of
7.23 percent in its analysis.

19

Table ES.3.22 External Power Supply Markups by Product Class


Composite
Output
Class ID
Baseline
Markup

Composite
Incremental
Markup

Class A

AC-DC

A1: 0-10.25 W

2.16

1.40

A1: 10.25-39 W

2.00

1.31

A1: 39-90 W

2.00

1.34

A1: 120-250 W

1.89

1.31

Low Voltage

A2

2.19

1.47

Basic Voltage

A3

2.07

1.31

Low Voltage

A4

2.17

1.33

100 W

X1

1.99

1.31

100 W

X2

2.27

1.45

>250 W

H1

1.69

1.16

Basic Voltage

M1

2.53

1.48

Low Voltage

M2

Basic Voltage

AC-AC
Non-Class A
Multiple-Voltage
High Power
AC-DC
For Medical
Devices
AC-AC
For MotorOperated BC
Applications
and Detachable
Batteries
(MADB)

AC-DC

AC-AC

Basic Voltage

2.53

1.48

M3

Low Voltage

M4

Basic Voltage

B1

2.16

1.35

Low Voltage

B2

2.01

1.23

Basic Voltage

B3*

Low Voltage

B4*

DOE did not identify products in product classes M3, M4, B3, and B4.

20

Table ES.3.23 Battery Charger Markups by Product Class


InputOutput

Battery
Energy

Class
ID

Voltage

<100 Wh
AC-DC
1003000 Wh

AC-AC

Composite
Incremental
Markup

Inductive Connection

2.00

1.23

<4 V

2.13

1.40

410 V

2.17

1.33

>10 V

1.98

1.32

<20 V

2.32

1.37

20 V

2.28

1.36

2.31

1.37

<9 V Input

2.12

1.38

9 V Input

2.60

1.69

AC Output from Battery

10

2.01

1.35

>3000 Wh
DC-DC

Composite
Baseline
Markup

ES.3.6 Energy Use Analysis


The energy use analysis assesses the potential energy use of higher efficiency products
and provides the basis for estimating the energy savings used in the LCC and subsequent
analyses. The DOE BC and EPS test procedures provide standardized results for each mode of
operation and can serve as the basis for comparing the performance of different appliances under
the same conditions. However, the way a given BC or EPS is used is also an important
determinant of its overall energy consumption.
To determine the variation in field energy use by products that would meet possible
energy efficiency standards, DOE assigned usage profiles to each BC and EPS application. DOE
derived these application-specific usage profiles based on comments by interested parties,
publications, and its own estimates.
To estimate EPS energy use DOE multiplied annual usage in each application state and
operating mode (in hours per year) by the power estimates (in watts). These power estimates are
a function of EPS efficiency and loading. Table ES.3.24 shows DOEs estimates of the annual
unit energy consumption of each representative unit at each CSL.
Table ES.3.24 Class A External Power Supply Unit Energy Consumption (kWh/yr)
Representative
Unit
2.5 W
18 W
60 W
120 W

CSL 0

CSL 1

CSL 2

CSL 3

CSL 4

3.6
15.5
11.7
19.4

2.5
11.8
10.3
16.8

1.9
9.7
8.7
14.2

1.7
7.9
7.3
13.6

1.6
6.6
5.4
9.4

21

To determine BC energy usage in the field, DOE multiplied annual usage in each
operating mode (in hours per day or charges per day) by the power estimates (in watts) or energy
per charge (in watt hours) from the engineering analysis. The usage profiles DOE developed for
each product class are shown in Table ES.3.25.
Table ES.3.25 Battery Charger Usage Profiles
Product
Active +
No Battery
Unplugged
Class ID
Maintenance
(Standby)
Hours per Day
1
23.9
0.1
0
2
9.7
5.0
9.4
3
5.6
0.2
18.1
4
19.8
0.3
6.9
5
7.7
0.5
15.8
6
15.4
8.6
0
7
7.7
8.1
8.1
8
6.5
7.5
10.1
9
1.1
0.1
22.8
10
24.0
0
0

Off

Charges
0
0
0.1
0.1
0
0
0
0
0
0

No. per Day


0.26
0.56
0.22
0.88
0.55
0.46
0.36
0.43
0.15
0

Note: Usage times may not sum to 24 hours due to rounding.

DOE estimated the annual unit energy consumption of BCs in each product class at each
CSL. Results are displayed in Table ES.3.26 and Table ES.3.27. These values are used in the
NIA and are the shipment-weighted averages of the application-specific UEC values used in the
LCC analysis. Chapter 7 provides more details on the methods, data, and assumptions employed
in the energy use analysis.
Table ES.3.26 Battery Charger Unit Energy Consumption Based on ManufacturerSupplied Data (kWh/yr)
Class ID
CSL 0
CSL 1
CSL 2
CSL 3
1
10.0
7.0
3.5
1.5
2
10.0
6.0
1.2
3
10.0
5.4
1.0
4
39.4
10.5
6.1
5
202.3
158.4
100.0
75.0
6
69.3
41.9
25.0
18.0
7
290.0
250.0
200.0
150.0
8
1.5
0.9
0.6
9
1.3
0.8
0.5
10
19.6
6.4
4.0
1.5

22

Table ES.3.27 Battery Charger Unit Energy Consumption based on Test and Teardown
Data (kWh/yr)
Class ID
CSL 0
CSL 1
CSL 2
CSL 3
1
2
7.4
1.2
3
4
39.4
10.5
6.1
5
232.7
179.3
6
7
8
9
10
19.6
6.4
ES.3.7 Life-Cycle Cost and Payback Period Analyses
The LCC and PBP analysis is used to determine the economic impact of potential
standards on end users. The effects of standards on individual end users include changes in
operating expenses (usually lower) and changes in total installed cost (usually higher).
DOE analyzed the net effect of these changes for BCs and EPSs, first by calculating the
changes in end users LCCs likely to result from various candidate standard levels compared to a
base case (no new standards). The LCC calculation considers total installed cost (which includes
manufacturer selling price, sales taxes, and distribution chain markups), operating expenses
(energy expenses), product lifetime, and discount rate. Because DOE found no evidence that
installation, repair, or maintenance costs apply to BCs or EPSs, it excluded those costs from its
analysis. DOE performed the LCC analysis from the perspective of users of BCs and EPSs.
The PBP is the amount of time it takes the customer to recover the assumed additional
installed cost of more efficient products through energy cost savings. Payback analysis is a
common technique used to evaluate investment decisions. This calculation is known as a simple
payback period because it does not account for changes in operating expense over time or the
time value of money.
The inputs to the PBP calculation are the total installed cost of the product to the end user
for each efficiency level and the first years operating expenditures for each efficiency level. The
PBP calculation uses the same inputs as the LCC analysis, except that energy price trends and
discount rates are not needed. The calculation uses energy prices only for the year of the
compliance date, in this case 2013.
To accurately estimate the share of consumers that would be affected by a standard at a
particular efficiency level, DOEs LCC analysis considered the projected distribution of product
efficiencies that consumers purchase under the base case (i.e., the case without new energy
efficiency standards). DOE refers to this distribution of product efficiencies as a base-case
efficiency distribution. DOE developed base-case efficiency distributions for each of the

23

representative product classes as projected for the year that new standards are assumed to
become effective (2013). Using the projected distribution of product efficiencies for each
representative unit, DOE determined the impact of each standard level by sampling consumers
assigned to each efficiency level in the base case. If a consumer was assigned a product
efficiency that was greater than or equal to the efficiency of the standard level under
consideration, the LCC calculation would show that the consumer was not impacted by that
standard level.
Table ES.3.28 through Table ES.3.41 show the results of the LCC and PBP analyses for
the representative units and representative product classes at each considered energy efficiency
level. Table ES.3.28, Table ES.3.29, Table ES.3.30, and Table ES.3.31 present the EPS results.
Table ES.3.32, Table ES.3.33, Table ES.3.34, Table ES.3.35, Table ES.3.36, Table
ES.3.37, Table ES.3.38, Table ES.3.39, Table ES.3.40, and Table ES.3.41 present the BC results.
Consumers are said to have no impact if the base case forecast product assigned to them has a
greater efficiency than the level indicated by a standard. In the case of the PBP, a very small
change in operating cost can result in a few extremely large payback periods, which will skew
the mean payback. In these cases, the median is a better indicator of payback period because it is
the value at the center of the distribution of results, which is unaffected by results that would be
considered outliers.
Figures presented in Chapter 8 show the range of LCC savings and PBPs for all of the
efficiency levels considered for each representative unit. Chapter 8 provides more details on the
methods, data, and assumptions used for the LCC and PBP analyses.
Table ES.3.28 LCC and PBP Results for 2.5W Regular AC/DC External Power Supplies
(Manufacturer Data)
Life-Cycle Cost
No
Avg.
Avg.
Load
Installed Operating
Eff.* Power Price
Cost
CSL
%
W
2009$
2009$
0
58.3
0.500
0.00
1.73
1
67.9
0.300
0.06
1.12
2
71.0
0.130
0.35
0.90
3
73.5
0.100
0.47
0.78
4
74.0
0.053
0.63
0.74
* Eff. stands for efficiency level.
Based on an incremental MSP over the baseline.

Avg.
LCC
2009$
1.73
1.19
1.24
1.25
1.36

24

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings Cost Impact Benefit
2009$
%
%
%
0.19
3.1
66.0
30.9
0.14 40.5
15.3
44.2
0.13 44.2
3.2
52.6
0.02 64.3
0.0
35.7

Payback Period

Median
yrs.
0.4
3.6
3.7
4.8

Mean
yrs.
1.3
12.7
12.7
17.2

Table ES.3.29 LCC and PBP Results for 18W Regular AC/DC External Power Supplies
(Manufacturer Data)
Life-Cycle Cost
No
Avg.
Avg.
Load
Installed Operating
Eff.* Power Price
Cost
CSL
%
W
2009$
2009$
0
76.0
0.500
0.00
6.54
1
80.3
0.300
0.44
4.99
2
83.0
0.200
0.59
4.11
3
85.4
0.100
1.11
3.35
4
87.5
0.062
1.73
2.79
* Eff. stands for efficiency level.
Based on an incremental MSP over the baseline.

Avg.
LCC
2009$
6.54
5.43
4.70
4.45
4.52

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings Cost Impact Benefit
2009$
%
%
%
0.33
6.2
65.9
27.9
0.96 11.5
15.3
73.2
1.20 18.2
3.3
78.5
1.13 33.5
0.0
66.5

Payback Period

Median
yrs.
1.1
0.8
1.5
2.1

Mean
yrs.
4.9
2.8
5.5
8.0

Table ES.3.30 LCC and PBP Results for 60W Regular AC/DC External Power Supplies
(Manufacturer Data)
Life-Cycle Cost
No
Avg.
Avg.
Load
Installed Operating
Eff.* Power Price
Cost
CSL
%
W
2009$
2009$
0
85.0
0.500
0.00
4.63
1
87.0
0.500
0.34
4.08
2
87.0
0.200
1.02
3.43
3
88.0
0.073
2.28
2.90
4
91.0
0.073
4.18
2.13
* Eff. stands for efficiency level.
Based on an incremental MSP over the baseline.

Avg.
LCC
2009$
4.63
4.43
4.46
5.18
6.31

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Savings Cost
Impact
2009$
%
%
0.07
4.8
66.0
0.05
50.0
15.3
-0.65
88.5
3.3
-1.78
96.7
0.0

Payback Period
Net
Benefit
%
29.2
34.7
8.2
3.3

Median
yrs.
2.4
3.6
5.8
7.2

Mean
yrs.
2.4
3.6
6.0
7.2

Table ES.3.31 LCC and PBP Results for 120W Regular AC/DC External Power Supplies
(Manufacturer Data)
Life-Cycle Cost
No
Avg.
Avg.
Load
Installed Operating
Eff.* Power Price
Cost
CSL
%
W
2009$
2009$
0
85.0
0.500
0.00
6.48
1
87.0
0.500
0.92
5.62
2
88.0
0.230
1.72
4.73
3
88.4
0.210
1.97
4.53
4
91.7
0.165
7.05
3.15
* Eff. stands for efficiency level.
Based on an incremental MSP over the baseline.

Avg.
LCC
2009$
6.48
6.54
6.45
6.50
10.20

25

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Savings Cost
Impact
2009$
%
%
-0.02
24.7
66.1
0.05
44.7
15.2
0.01
63.1
3.2
-3.69
99.8
0.0

Payback Period
Net
Benefit
%
9.2
40.1
33.7
0.2

Median
yrs.
3.6
3.0
3.3
7.6

Mean
yrs.
3.4
2.9
3.2
7.8

Table ES.3.32 LCC and PBP Results for Product Class 1: Low Energy, Inductive Battery
Chargers (Manufacturer Data)

CSL
0
1
2
3

UEC
kWh
10.0
7.0
3.5
1.5

Life-Cycle Cost
Avg.
Avg.
Installed Operating
Price
Cost
2009$
2009$
4.39
4.81
4.61
3.37
4.91
1.68
5.12
0.72

Avg.
LCC
2009$
9.20
7.97
6.59
5.84

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Savings
Cost
Impact
2009$
%
%
0.93
0.0
24.9
2.14
0.0
12.4
2.90
0.0
0.0

Payback Period
Net
Benefit
%
75.1
87.6
100.0

Median
yrs.
0.7
0.8
0.9

Mean
yrs.
0.7
0.8
0.8

Table ES.3.33 LCC and PBP Results for Product Class 2: Low Energy, Low Voltage
Battery Chargers (Manufacturer Data)

CSL
0
1
2

UEC
kWh
10.0
6.0
1.2

Life-Cycle Cost
Avg.
Avg.
Installed Operating
Price
Cost
2009$
2009$
1.42
3.30
2.13
2.08
18.09
0.46

Avg.
LCC
2009$
4.72
4.20
18.54

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Savings
Cost
Impact
2009$
%
%
0.20
6.6
60.8
-13.74
97.2
2.8

Payback Period
Net
Benefit
%
32.6
0.0

Median
yrs.
1.5
25.4

Mean
yrs.
2.2
38.9

Table ES.3.34 LCC and PBP Results for Product Class 2: Low Energy, Low Voltage
Battery Chargers (Test/Teardown Data)

CSL
0
1
2

UEC
kWh
10.0
6.0
1.2

Avg.
Installed
Price
2009$
N/A
1.42
4.43

Life-Cycle Cost
Avg.
Operating
Avg.
Cost
LCC
2009$
2009$
N/A
N/A
2.48
3.90
0.38
4.82

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings
Cost
Impact
Benefit
2009$
%
%
%
N/A
N/A
N/A
N/A
-0.90
73.7
2.8
23.5

Payback Period

Median
yrs.
N/A
4.7

Mean
yrs.
N/A
6.7

Table ES.3.35 LCC and PBP Results for Product Class 4: Low Energy, High Voltage
Battery Chargers (Manufacturer Data)

CSL
0
1
2

UEC
kWh
39.4
10.5
6.1

Avg.
Installed
Price
2009$
8.04
16.13
20.60

Life-Cycle Cost
Avg.
Operating
Avg.
Cost
LCC
2009$
2009$
16.65
24.69
4.31
20.44
2.48
23.08

Life-Cycle Cost Savings


Consumers with
Net
No
Net
Cost
Impact
Benefit
%
%
%
0.97
9.0
81.3
9.7
-0.65
53.8
38.3
7.9

Wtd.
Avg.
Savings
2009$

26

Payback Period

Median
yrs.
3.4
9.3

Mean
yrs.
3.1
9.1

Table ES.3.36 LCC and PBP Results for Product Class 4: Low Energy, High Voltage
Battery Chargers (Test/Teardown)

CSL
0
1
2

UEC
kWh
39.4
10.5
6.1

Avg.
Installed
Price
2009$
8.04
12.23
13.20

Life-Cycle Cost
Avg.
Operating
Avg.
Cost
LCC
2009$
2009$
15.38
23.42
4.14
16.38
2.52
15.72

Life-Cycle Cost Savings


Consumers with
Net
No
Net
Cost
Impact
Benefit
%
%
%
1.33
0.4
81.5
18.1
1.74
11.1
38.2
50.7

Wtd.
Avg.
Savings
2009$

Payback Period

Median
yrs.
1.3
1.5

Mean
yrs.
1.6
2.5

Table ES.3.37 LCC and PBP Results for Product Class 5: Medium Energy, Low Voltage
Battery Chargers (Manufacturer Data)

CSL
0
1
2
3

UEC
kWh
202.7
159.6
100.0
75.0

Life-Cycle Cost
Avg.
Avg.
Installed Operating
Price
Cost
2009$
2009$
45.91
90.58
59.09
72.76
112.93
40.69
205.20
30.43

Avg.
LCC
2009$
136.49
131.85
153.62
235.63

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Savings
Cost
Impact
2009$
%
%
2.08
10.0
56.9
-11.07
48.4
40.5
-93.07
99.9
0.0

Payback Period
Net
Benefit
%
33.1
11.1
0.1

Median
yrs.
2.4
6.4
17.2

Mean
yrs.
8.1
7.7
42.2

Table ES.3.38 LCC and PBP Results for Product Class 5: Medium Energy, Low Voltage
Battery Chargers (Test/Teardown Data)

CSL
0
1

UEC
kWh
202.7
159.6

Avg.
Installed
Price
2009$
45.91
50.64

Life-Cycle Cost
Avg.
Operating
Avg.
Cost
LCC
2009$
2009$
112.90
158.81
83.11
133.75

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings
Cost
Impact
Benefit
2009$
%
%
%
11.20
0.0
56.2
43.8

Payback Period

Median
yrs.
0.9

Mean
yrs.
0.9

Table ES.3.39 LCC and PBP Results for Product Class 7: High Energy Battery Chargers
(Manufacturer Data)

CSL
0
1
2
3

UEC
kWh
290.0
250.0
200.0
150.0

Avg.
Installed
Price
2009$
195.80
219.13
266.16
361.77

Life-Cycle Cost
Avg.
Operating
Avg.
Cost
LCC
2009$
2009$
171.85
367.65
148.15
367.28
118.52
384.68
88.89
450.66

Life-Cycle Cost Savings


Consumers with
Net
No
Net
Cost
Impact
Benefit
%
%
%
0.10
29.7
50.0
20.3
-12.07
64.2
30.1
5.7
-78.04 100.0
0.0
0.0

Wtd.
Avg.
Savings
2009$

27

Payback Period

Median
yrs.
5.8
7.8
12.6

Mean
yrs.
5.4
7.7
13.6

Table ES.3.40 LCC and PBP Results for Product Class 10: Low Energy, AC Out Battery
Chargers (Manufacturer Data)

CSL
0
1
2
3

UEC
kWh
19.6
6.4
4.0
1.5

Avg.
Installed
Price
2009$
5.93
7.62
8.08
8.65

Life-Cycle Cost
Avg.
Operating
Avg.
Cost
LCC
2009$
2009$
11.07
17.00
3.60
11.22
2.26
10.34
0.85
9.50

Life-Cycle Cost Savings


Consumers with
Net
No
Net
Cost
Impact
Benefit
%
%
%
5.16
0.0
11.0
89.0
6.04
0.0
0.0
100.0
6.88
0.0
0.0
100.0

Wtd.
Avg.
Savings
2009$

Payback Period

Median
yrs.
1.3
1.4
1.5

Mean
yrs.
1.2
1.3
1.5

Table ES.3.41 LCC and PBP Results for Product Class 10: Low Energy, AC Out Battery
Chargers (Test/Teardown Data)

CSL
0
1

UEC
kWh
19.6
6.4

Avg.
Installed
Price
2009$
5.93
4.54

Life-Cycle Cost
Avg.
Operating
Avg.
Cost
LCC
2009$
2009$
11.07
17.00
3.60
8.14

Life-Cycle Cost Savings


Consumers with
Net
No
Net
Cost
Impact
Benefit
%
%
%
7.88
0.0
11.1
88.9

Wtd.
Avg.
Savings
2009$

Payback Period

Median
yrs.
0.0

Mean
yrs.
0.0

DOE conducted sensitivity analyses to quantify the relationship between certain critical
inputs to the life-cycle cost (LCC) analysis and the resultant weighted-average LCC savings for
EPSs. These critical inputs were MSPs, product lifetimes, usage profiles, and loading points. By
varying the critical inputs simultaneously, DOE calculated a low-savings case and a high-savings
case for weighted-average LCC savings to compare with the reference case previously presented.
The sensitivity analysis demonstrates that in the high-savings case all analyzed EPSs have at
least one CSL with positive LCC savings. Table ES.3.42 presents a summary of the EPS LCC
sensitivity analysis results, indicating whether LCC savings are positive or negative in
comparison to the reference case. Table ES.3.43 presents the detailed results of the EPS LCC
sensitivity analysis by CSL.

28

Table ES.3.42 EPS Weighted-Average LCC Savings Sensitivity Analysis Summary

2.5 W

18 W

60 W

120 W

CSL 1
CSL 2
CSL 3
CSL 4
CSL 1
CSL 2
CSL 3
CSL 4
CSL 1
CSL 2
CSL 3
CSL 4
CSL 1
CSL 2
CSL 3
CSL 4

Reference Case
Positive
Positive
Positive
Positive
Positive
Positive
Positive
Positive
Positive
Positive
Negative
Negative
Negative
Positive
Positive
Negative

High Savings Case

Low Savings Case

No changes in sign
(All CSLs are positive)

CSL 4 is now negative


No other changes in sign

No changes in sign
(All CSLs are positive)

CSL 4 is now negative


No other changes in sign

No changes in sign

CSL 2 is now negative


No other changes in sign

CSL 1 is now positive


No other changes in sign

All CSLs are negative

Table ES.3.43 EPS Weighted-Average LCC Savings by CSL

2.5 W

18 W

60 W

120 W

CSL 1
CSL 2
CSL 3
CSL 4
CSL 1
CSL 2
CSL 3
CSL 4
CSL 1
CSL 2
CSL 3
CSL 4
CSL 1
CSL 2
CSL 3
CSL 4

Reference
Case
2009$
0.19
0.14
0.13
0.02
0.33
0.96
1.20
1.13
0.07
0.05
(0.65)
(1.78)
(0.02)
0.05
0.01
(3.69)

High Savings
Case
2009$
0.19
0.23
0.25
0.17
0.51
1.27
1.79
2.04
0.12
0.27
(0.13)
(0.78)
0.16
0.63
0.72
(1.39)

Low Savings
Case
2009$
015
0.08
0.06
(0.06)
0.21
0.65
0.42
(0.19)
0.00
(0.26)
(1.42)
(3.30)
(0.15)
(0.37)
(0.51)
(5.72)

DOE also conducted sensitivity analyses in a manner similar to that it used for EPSs to
quantify the relationship between certain critical inputs to the LCC analysis and the resultant
weighted-average LCC savings for BCs. These critical inputs were MSPs, product lifetimes, and
usage profiles. As explained earlier, DOE analyzed LCC results for both manufacturer data and
test and teardown data. DOE adjusted MSPs, product lifetimes, and usage profiles for the
manufacturer data. For the teardown results, DOE varied all the inputs except for the MSP,

29

because the MSP was based on iSupplia teardowns of physical units, which did not have an
associated uncertainty. By varying the critical inputs simultaneously, DOE calculated a lowsavings case and a high-savings case for weighted-average LCC savings to compare with the
reference case presented above. Table ES.3.44 presents a summary of the BC LCC sensitivity
analysis results, indicating whether LCC savings are positive or negative in comparison to the
reference case. Table ES.3.45 presents the detailed results of the BC LCC sensitivity analysis by
product class.
Table ES.3.44 BC Weighted-Average LCC Savings Sensitivity Analysis Summary

PC 1

PC 2

PC 4

PC 5

PC 7

PC 10

CSL 1

Reference
Case
Positive

CSL 2

Positive

CSL 3

Positive

CSL 1

Positive

CSL 2

Negative

CSL 3

N/A

CSL 1

Positive

CSL 2

Negative

CSL 3

N/A

CSL 1

Positive

CSL 2

Negative

CSL 3

Negative

CSL 1

Positive

CSL 2

Negative

CSL 3

Negative

CSL 1

Positive

CSL 2

Positive

CSL 3

Positive

Manufacturer Data
High Savings
Low Savings
Case
Case
No changes in
No changes in
sign
sign
(All CSLs are
(All CSLs are
positive)
positive)

Test and Teardown Data


Reference
High Savings Low Savings
Case
Case
Case
No Data

No changes in
sign

No changes in
sign

N/A
Negative
N/A

All CSLs are


positive

All CSLs are


negative

Positive
Positive
N/A

No changes in
sign

All CSLs are


negative

Positive
N/A
N/A

CSL 1 and
CSL 2 are
positive

All CSLs are


negative

No changes in
sign
(All CSLs are
positive)

No changes in
sign
(All CSLs are
positive)

No changes in
sign
(All CSLs are
negative)
No changes in
sign
(All CSLs are
positive)
No changes in
sign
(All CSLs are
positive)

No changes
in sign
(All CSLs are
negative)
No changes
in sign
(All CSLs are
positive)
No changes
in sign
(All CSLs are
positive)

No Data

Positive
N/A
N/A

No changes in
sign
(All CSLs are
positive)

No changes
in sign
(All CSLs are
positive)

iSuppli is a DOE subcontractor tasked with disassembling various battery chargers and external power supplies and
developing a high volume cost for the individual circuit elements.

30

Table ES.3.45 BC Weighted-Average LCC Savings by Product Class

PC 1: Low E,
Inductive

PC 2: Low E,
Low V

PC 4: Low E,
High V

PC 5: Medium E,
Low V

PC 7: High E

PC 10: Low E,
AC Out

CSL 1

Manufacturer Data (2009$)


Reference
High
Low
Case
Case
Case
0.93
0.94
0.89

Test and Teardown Data (2009$)


Reference
High
Low
Case
Case
Case

CSL 2

2.14

2.19

2.05

CSL 3

2.90

2.98

2.78

CSL 1

0.20

0.31

0.08

N/A

N/A

N/A

CSL 2

(13.74)

(11.47)

(17.74)

(0.90)

(0.61)

(1.21)

CSL 3

N/A

N/A

N/A

N/A

N/A

N/A

CSL 1

0.97

1.86

(0.23)

1.33

1.95

1.03

CSL 2

(0.65)

1.07

(2.83)

1.74

2.64

1.35

CSL 3

N/A

N/A

N/A

N/A

N/A

N/A

CSL 1

2.08

2.99

(4.82)

11.20

N/A*

7.58

CSL 2

(11.07)

(6.49)

(29.68)

N/A

N/A

N/A

CSL 3

(93.07)

(77.74)

(136.75)

N/A

N/A

N/A

CSL 1

0.10

5.93

(6.83)

CSL 2

(12.07)

5.03

(32.69)

CSL 3

(41.27)

(122.34)

CSL 1

(78.04)
5.16

5.34

4.93

7.88

N/A*

N/A*

CSL 2

6.04

6.28

5.74

N/A

N/A

N/A

N/A

N/A

N/A

6.88
7.19
6.48
CSL 3
*
Inputs did not change from the reference case. See reference case data.

No Data

No Data

Appendix 8B provides more details on the sensitivity analyses methodology and results.
ES.3.8 Shipments Analysis
The shipments analysis generates shipment forecasts and base case efficiency
distributions for each of the product classes being considered in this preliminary analysis of
standards for battery chargers and external power supplies. These outputs allow DOE to assess
the scale of savings potential, in terms of the number of units that could be affected by potential
standards. Outputs from the shipments analysis are inputs to the life-cycle cost analysis, national
impact analysis, and manufacturer impact analysis.
Chapter 9 provides more details on the methods, data, and assumptions used for the
shipments analysis.
ES.3.9 National Impact Analysis
The national impact analysis (NIA) assesses the following impacts from potential energy
efficiency standards:

31

(1) the national energy savings (NES) attributable to each possible standard, calculated
by combining unit energy savings (from the energy use analysis) with the number of units
affected by a potential standard, which is itself a combination of the products estimated lifetime
(from the market assessment), shipments and efficiency distributions (from the shipments
analysis). Initial savings are calculated in site kWh, and then converted to source Btu.
(2) the monetary value of those energy savings to consumers of the considered products,
calculated by combining NES with forecast electricity prices.
(3) the increased cost of the products because of standards, calculated by combining the
unit improvement costs with the number of units affected by a standard, and
(4) net present value (NPV) of total consumer costs and savings, calculated as the
difference between the value of energy savings less increased product costs, discounted to the
present.

In order to determine the influence of inputs on the NIA, DOE conducted an NIA
sensitivity analysis that mirrored the LCC sensitivity analysis. The results of the NIA sensitivity
analysis are presented below and in chapter 10.
To give some sense for the relative significance of various types of BCs and EPSs, DOE
estimated the national aggregate annual energy consumption of products in several market
segments. These estimates are shown in Table ES.3.46 for EPSs and Table ES.3.47 for BCs.
When evaluated in the reference savings case, the EPSs and BCs that shipped in 2008 consume
an estimated 2.2 billion kWh/yr and 3.4 billion kWh/yr of electricity, respectively.
Table ES.3.46 Estimated National Annual Energy Consumption of External Power
Supplies Shipped in 2008
Annual Energy Consumption
Product Class

Site Energy Primary Energy


(Million kWh) (Trillion Btu)

Percent of
Total

A1: 0-10.25 W

251

11%

A1: 10.25-39 W

930

10

42%

A1: 39-90 W

582

26%

A1: 90-250 W

186

8%

A2 (DC Output, Low Voltage)

144

6%

A3 (AC Output, Basic Voltage)

113

5%

A4 (AC Output, Low Voltage)

17

1%

2,222

24

100%

All Class A EPSs

Note: Figures calculated using DOEs estimates of the efficiency of products in the market in 2009.
Individual values may not sum to totals due to rounding.

32

Table ES.3.47 Estimated National Annual Energy Consumption of Battery Chargers


Shipped in 2008
Conversion Battery
Annual Energy Consumption
Type
Energy
Battery
Class
Site Energy
Primary
Percent
Voltage
ID
(Million
Energy
of Total
kWh)
(Trillion Btu)

AC-DC

<100 Wh

1003,000 Wh

Inductive
Connection

<4 V

47

1%

1,539

17

45%

4<10 V

148

4%

10 V

991

11

29%

<20 V

359

11%

20 V

48

1%

55

2%

<9 V Input

80

2%

9 V Input

15

0.2

<1%

AC Output
from
Battery

10
125

4%

3,408

37

100%

>3,000 Wh
DC-DC

AC-AC
All Battery Chargers

Note: Figures calculated using DOEs estimates of the efficiency of products in the market in 2009.
Individual values may not sum to totals due to rounding.

Chapter 10 provides more details on the methods, data, and assumptions used for the
national impact analysis.
ES.3.9.1

National Energy Savings Analysis

DOEs preliminary analysis finds that new standards for BCs and amended standards for
EPSs could produce cumulative energy savings over the period of analysis of as much as
3.5 quadrillion Btu (quads) and 1.5 quads, respectively, in the case where energy savings are
maximized. In the case where consumer benefits are maximized, new standards for BCs and
amended standards for EPSs could produce cumulative energy savings over the period of
analysis of 1.4 quads and 1.0 quads, respectively. These totals assume reference case input
values and manufacturer-derived cost data.
Table ES.3.48, Table ES.3.49 and Table ES.3.50 show DOEs estimates of national
energy savings potential at each CSL for EPS and BC product classes. In each cell, results from
the reference case are shown in the top of the cell, with results from the low and high cases
shown in parentheses in the bottom of the cell. Savings in the low and high savings cases were
only calculated for the representative product classes: EPS product class A1 and BC product

33

classes 2, 4, 5, 7, and 10. Because of the unavailability of data regarding the other classes, DOE
took this more limited approach in its analysis.
Table ES.3.48 National Energy Savings Potential from Amended Standards for Class A
External Power Supplies (Trillion Btu)
Standard Level Selected
Product Class
CSL 1
CSL 2
CSL 3
CSL 4
57
114
144
158
A1: 0-10.25 W
(46 to 58)
(99 to 142) (127 to 180) (140 to 202)
132
319
504
643
A1: 10.25-39 W
(119 to 145) (287 to 352) (453 to 555) (578 to 710)
34
137
232
374
A1: 39-90 W
(31 to 43)
(128 to 167) (217 to 286) (345 to 466)
10
35
42
88
A1: 90-250 W
(9 to 15)
(32 to 54)
(38 to 64)
(78 to 137)
32
81
96
111
A2 (DC Output, Low Voltage)
A3 (AC Output, Basic Voltage)

26

64

101

128

A4 (AC Output, Low Voltage)

296
(267 to 325)

756
(697 to 866)

9
1128
(1041 to
1291)

10
1512
(1389 to
1763)

All Class A EPSs

Note: Individual values may not sum to totals due to rounding.

34

Table ES.3.49 National Energy Savings Potential from New Standards for Battery
Chargers (Trillion Btu), Manufacturer Data
Conversion Battery
Standard Level Selected
Class
Type
Energy Battery
Voltage
ID
CSL 1
CSL 2
CSL 3
Inductive
22
52
71
1
Connection
<100
395
1591
Wh
<4 V
2
N/A
AC-DC
(337 to 421)
(1271 to 1785)
59

206

541
(465 to 764)
88
(19 to 88)

810
(692 to 1144)
248
(109 to 248)

364
(135 to 364)

41

76

97

9
(4 to 15)

26
(12 to 42)

49
(24 to 81)

<9 V Input

21

50

N/A

9 V Input

14

N/A

AC Output
from
Battery

10

176

212

249

4<10 V

10 V

100<20 V
3000 Wh

20 V
>3000
Wh
DC-DC

AC-AC

35

N/A
N/A

Table ES.3.50 National Energy Savings Potential from New Standards for Battery
Chargers (Trillion Btu), Test and Teardown Data
Conversion Battery
Standard Level Selected
Class
Type
Energy Battery
Voltage
ID
CSL 1
CSL 2
CSL 3
Inductive
No Data
1
Connection
<100
1536
Wh
<4 V
2
N/A
N/A
(1268 to 1727)
AC-DC
100
3000
Wh

10 V

<20 V

20 V

No Data

No Data

<9 V Input

No Data

9 V Input

No Data

AC Output
from
Battery

10

>3000
Wh
DC-DC

AC-AC

ES.3.9.2

No Data

4<10 V

542
(476 to 770)
107
(63 to 107)

825
(715 to 1053)

812
(720 to 1159)

N/A

N/A

N/A

N/A

N/A

Net Present Value Analysis

DOEs preliminary analysis finds that new standards for BCs and amended standards for
EPSs would yield net consumer savings over the period of analysis of $4.5 billion and
$3.4 billion, respectively, in the case where a CSL is selected for each product class to maximize
consumer benefits. In the case where energy savings are maximized, standards would yield net
consumer savings over the period of analysis of -$70 billion and $0.25 billion for BCs and EPSs,
respectively. These totals assume reference case input values and manufacturer-derived cost
data.
Table ES.3.51, Table ES.3.52, Table ES.3.53, Table ES.3.54, Table ES.3.55 and Table
ES.3.56 show the NPV results for the EPS and BC product classes at each CSL. In each cell,
results from the reference case are shown in the top of the cell, with results from the low and
high cases shown in parentheses in the bottom of the cell. Savings in the low and high savings
cases were only calculated for the representative product classes: EPS product class A1 and BC
product classes 1, 2, 4, 5, 7, and 10. As indicated earlier, data limitations restricted DOEs ability
to calculate parallel savings from other classes.

36

Table ES.3.51 Net Present Value of Amended Standards for Class A External Power
Supplies at a 3 Percent Discount Rate (2009$ millions)
Product Class
Standard Level Selected
CSL 1
CSL 2
CSL 3
CSL 4
332
315
322
A1: 0-10.25 W
(261 to
(212 to
(202 to
160
335)
496)
562)
(29 to 457)
637
1663
2155
2186
A1: 10.25-39 W
(391 to
(1164 to
(1007 to
(281 to
807)
2027)
2921)
3402)
117
-606
-1785
A1: 39-90 W
90
(-199 to
(-1426 to
(-3415 to (24 to 172)
442)
96)
476)
-768
A1: 90-250 W
-2
18
10
(-1157 to (-27 to 47) (-56 to 178) (-81 to 205)
183)
A2 (DC Output, Low
Voltage)
174
106
22
-136
A3 (AC Output, Basic
Voltage)
138
348
492
560
A4 (AC Output, Low
Voltage)
26
32
42
38
Table ES.3.52 Net Present Value of Amended Standards for External Power Supplies at a
7 Percent Discount Rate (2009$ millions)
Product Class
Standard Level Selected
CSL 1
CSL 2
CSL 3
CSL 4
165
A1: 0-10.25 W
(130 to
138
133
40
164)
(88 to 224) (75 to 247) (-24 to 183)
311
819
1035
1012
A1: 10.25-39 W
(180 to
(556 to
(422 to
(-10 to
400)
1008)
1439)
1658)
29
-392
-1074
A1: 39-90 W
41
(-142 to
(-836 to (-1957 to (6 to 82)
195)
29)
393)
-438
A1: 90-250 W
-3
3
-3
(-649 to (-16 to 22) (-37 to 84) (-52 to 96)
136)
A2 (DC Output, Low
Voltage)
86
37
-14
-104
A3 (AC Output, Basic
Voltage)
63
161
223
248
A4 (AC Output, Low
Voltage)
13
16
20
18

37

Table ES.3.53 Net Present Value of New Standards for Battery Chargers at a 3 Percent
Discount Rate (2009$ millions), Manufacturer Data
Standard Level Selected
Conversion Battery Battery
Class
Type
Energy Voltage
ID
CSL 1
CSL 2
CSL 3

<100
Wh
AC-DC
100
3000
Wh

Inductive
Connection

121
(118 to 123)

<4 V

1360
(718 to 1654)

4<10 V

10 V

<20 V

20 V

AC-AC

382
(369 to 391)

274
(-213 to 306)

-277
(-1503 to -64)
-603
-37
(-135 to 72)
-1970

-4026
(-6658 to
-3293)
-2277
-315
(-525 to -90)
N/A

N/A
N/A
N/A

<9 V Input

0.05
7
(-27 to 45)
-552

9 V Input

-80

-958

N/A

898
(857 to 923)

1056
(1003 to 1089)

1211
(1142 to
1253)

>3000 Wh
DC-DC

-39
1291
(265 to 2988)

281
(272 to 287)
-57441
(-74744 to
-48000)
-3957
-980
(-3283 to 2023)

AC Output
from
10
Battery

38

Table ES.3.54 Net Present Value of New Standards for Battery Chargers at a 7 Percent
Discount Rate (2009$ millions), Manufacturer Data
Standard Level Selected
Conversion Battery Battery
Class
Type
Energy Voltage
ID
CSL 1
CSL 2
CSL 3
Inductive
1
Connection
<100
Wh
AC-DC
100
3000
Wh

DC-DC

AC-AC

<4 V

4<10 V

10 V

<20 V

20 V

>3000 Wh

59
(57 to 60)

-40
555
(31 to 1361)

137
(132 to 140)
-31855
(-41224 to 26790)
-2,229
-762
(-1967 to 696)

115
(-124 to 132)

-251
(-866 to -134)
-370
-30
(-76 to 15)
-1,090

-2349
(-3699 to 1948)
-1,298
-191
(-292 to -96)
N/A

641
(310 to 783)

187
(179 to 191)
N/A
N/A
N/A

<9 V Input

-21
0.2
(-16 to 16)
-308

9 V Input

-45

-529

N/A

AC Output
from
Battery

10

425
(402 to 438)

498
(469 to 516)

569
(531 to 592)

39

Table ES.3.55 Net Present Value of New Standards for Battery Chargers at a 3 Percent
Discount Rate (2009$ millions), Test and Teardown Data
Standard Level Selected
Conversion Battery Battery
Class
Type
Energy Voltage
ID
CSL 1
CSL 2
CSL 3
Inductive
1
Connection
<100
Wh
AC-DC
100
3000
Wh

DC-DC

AC-AC

No Data
-2638
(-4342 to -1467)
No Data

<4 V

4<10 V

10 V

<20 V

20 V

No Data

>3000 Wh

No Data

<9 V Input

No Data

9 V Input

No Data

AC Output
from
Battery

10

N/A

2351
(1949 to
3744)
570
(293 to 570)

1232

N/A

3225
(2661 to 5345)

N/A

N/A

N/A

N/A

N/A

Table ES.3.56 Net Present Value of New Standards for Battery Chargers at a 7 Percent
Discount Rate (2009$ millions), Test and Teardown Data
Standard Level Selected
Conversion Battery Battery
Class
Type
Energy Voltage
ID
CSL 1
CSL 2
CSL 3
Inductive
1
Connection
<100
Wh
AC-DC
100
3000
Wh

DC-DC

No Data
N/A

-1845
(-2689 to -1304)

<4 V

4<10 V

10 V

<20 V

20 V

No Data

>3000 Wh

No Data

<9 V Input

No Data

9 V Input

No Data

N/A

No Data
1136
(950 to 1773)
269
(135 to 269)

40

1539
(1280 to 2509)

N/A

N/A

N/A

AC-AC

AC Output
from
Battery

10

608

N/A

N/A

ES.3.10 Preliminary Manufacturer Impact Analysis


A preliminary manufacturer impact analysis (MIA) assesses the potential impacts of
energy conservation standards on manufacturers, such as effects on expenditures for capital
conversion, marketing costs, shipments, and research and development costs. Potential impacts
might lead to changes in manufacturing practices for battery chargers and external power
supplies. DOE identified potential impacts through interviews with manufacturers and other
interested parties.
Manufacturers and other parties identified eight general areas of concern for external
power supplies and battery chargers. Relating to standards for external power supplies,
manufacturers and other parties identified four general areas of concern: (1) impacts on
shipments, (2) timing of standards, (3) efficiency levels, and (4) regulatory approval.
Relating to standards for battery chargers, manufacturers and other parties identified four
general areas of concern: (1) conversion costs, (2) efficiency as a selling point, (3) profit
margins, and (4) regulatory burden. See chapter 12 of the preliminary TSD for details.
ES.3.11 Other Analyses
The remaining chapters of the preliminary TSD address the analyses to be performed for the
notice of proposed rulemaking (NOPR):
The life-cycle cost analysis for subgroups evaluates the effects of energy conservation
standards on various national subgroups (chapter 11).
The utility impact analysis examines impacts of energy conservation standards on the
generation capacity of electric utilities (chapter 13).
The employment impact analysis examines the effects of energy conservation standards
on national employment (chapter 14).
The environmental assessment examines the effects of energy conservation standards on
various airborne emissions.
The regulatory impact analysis examines the national impacts of non-regulatory
alternatives to mandatory energy conservation standards.
ES.4

ISSUES ON WHICH DOE SEEKS PUBLIC COMMENT

DOE is interested in receiving comments on all aspects of this preliminary analysis. DOE
especially invites comments and data to improve this preliminary analysis, including data or
information that will respond to the following questions or concerns that were raised in response
to the framework document and in preparation of the preliminary TSD.

41

ES.4.1 Criteria for Identifying Charge Control in Wall Adapters


DOE is considering adopting the EPA criteria for establishing the presence of charge
control in a wall adapter, which are: (1) not having batteries or battery packs that physically
attach directly (including those that are removable) to the power supply unit or (2) not having a
battery chemistry or type selector switch AND an indicator light or state of charge meter (e.g., a
product with a type selector switch AND a state of charge meter is excluded from this
specification; a product with only an indicator light is still covered by this specification). DOE
welcomes comment on the appropriateness of these criteria. This issue is discussed in chapter 3
of the preliminary TSD.
ES.4.2 Design Options for Battery Chargers and External Power Supplies
DOE requests comment and data regarding additional design options or variants of the
considered design options that can increase the range of considered efficiency improvements for
either BCs or EPSs, including design options that may not yet be found in the market.
ES.4.3 Product Classes for Battery Chargers
DOE seeks comment on its battery charger product classes. In particular DOE seeks
comment on its decision to group BCs for both power tools and notebook computer into the same
low energy, high voltage product class.
ES.4.4 Dissimilarities Between Low Energy, High Voltage Battery Chargers
The low energy, high voltage product class includes both power tools and notebook
computers. Although these two products are shifting from nickel to lithium-ion batteries and
using more similar charge rates, DOE is still aware that differences between BCs for the two
products may exist. DOE welcomes comment on any dissimilarities between BCs for these two
types of low energy, high voltage products, and how best to address them in the analysis, e.g.,
through the creation of additional product classes.
ES.4.5 Scaling Within Battery Charger Product Classes
During interviews, some manufacturers suggested using the EPS ENERGY STAR
specification equations to relate output power to efficiency. Both BCs and EPSs are power
conversion circuits with the same general designs and components, and this method is already
being used in the EPS engineering analysis. DOE seeks comments on using the EPS ENERGY
STAR curve to scale the active-mode efficiency of BCs.
ES.4.6 Medium Energy, Low Voltage Representative Unit for BCs
DOE seeks comment on whether the marine charger is the most appropriate application
to focus its representative-unit analysis for the medium energy, low voltage product class.

42

ES.4.7 EPS Representative Units


DOE selected EPS representative units by focusing on popular units in the market, which
was straightforward for most representative units, except the low-power representative unit.
Although DOE considered a 5-watt representative unit because of its prevalence, DOE instead
selected a 2.5-watt representative unit because it belonged to the representative product class A1
whereas the 5-watt EPS was in the non-analyzed product class A2. DOE invites comment on this
approach in general and on whether to perform a detailed analysis for 5-watt, 5-volt EPSs in
particular. See chapter 5 of the preliminary TSD for more detail.
ES.4.8 EPS Max-Tech Value for 2.5-Watt Representative Unit
DOE verified the reasonableness of the aggregated manufacturer max-tech data which it
used to create curve fit equations for CSL 4 (max-tech). To that end, DOEs subject matter
experts (SMEs) reviewed the data and confirmed that the data fell within the expected ranges of
efficiencies based on their extensive experience with EPSs, other than the max-tech value for the
2.5W EPSs. The SMEs believe that 2.5W EPSs may be able to achieve a max tech efficiency of
80 percent rather than the 74.0 percent efficiency derived from manufacturers. DOE seeks
comment on the 2.5W max tech efficiency value used. DOE also seeks comment on the cost of
an 80 percent-efficient EPS, which it would need in order to use that as the max-tech efficiency
level for 2.5W EPSs in the NOPR.
ES.4.9 Efficiency Metrics for Multiple-Voltage EPSs
On April 2, 2010, DOE proposed a test procedure for multiple-voltage EPSs that would
yield five values: no-load power consumption as well as efficiency at 25 percent, 50 percent, 75
percent, and 100 percent of maximum load. 75 FR 16958. If DOE finalizes that proposal, DOE
would then propose to integrate the four efficiency values into an average efficiency metric in
the standard. Specifically, DOE proposes to use the no-load power consumption and an average
of the four efficiency values as metrics to set candidate standard levels for multiple-voltage
EPSs, similar to those set for single-voltage EPSs. Alternatively, DOE is also considering
approaches to integrate the four active mode measurements in a way that may better indicate the
typical efficiency of multiple-voltage EPSs during typical use. For instance, DOE could average
only the efficiency values at 50 percent and 75 percent, since those values are closest to the
operating conditions described by a stakeholder (Microsoft, EERE2008BTTP0004 No.8 at p.
1-2). Overall, DOE faces the same challenge with multiple-voltage EPSs as it does with singlevoltage EPSs: products change continuously, so the test metric should be robust enough to be
valid for all applicable products. DOE is seeking stakeholder comment on alternative methods
for using the five metrics to set candidate standard levels for EPSs.
ES.4.10 Representative Product Classes and Units for MADB EPSs
DOE chose to treat only MADB EPS product class B1 as representative because there are
significant similarities between the MADB product classes. Specifically, the MADB EPS
product classes have the same product class structure as the Class A EPS product classes, as
explained in chapter 3. To that end, DOE applied the same structure as Class A EPS for the
MADB EPS representative product classes and representative units. Since DOE can apply all

43

Class A EPS analysis to MADB EPSs, it intends also to include a 120-watt representative unit.
However, DOE has not identified an associated application for that unit. DOE is seeking
stakeholder comment on any applications that use a 120-watt MADB EPS and whether to
analyze that unit if it cannot identify any associated applications.
ES.4.11 Baseline Candidate Standard Level of Efficiency for Medical and MADB EPSs
In the determination analysis DOE defined the efficiency values for each CSL based on
existing efficiency levels in the market, test data and manufacturer data. For the NOPR, DOE is
considering gathering more data for medical and MADB EPSs with which to develop specific
efficiency values for the baseline CSL 0. This would allow DOE to characterize the benefits of
higher standards to consumers purchasing EPSs with efficiencies below EISA requirements.
However, DOE believes that consumers of medical and MADB EPSs with efficiencies below
EISA constitute a small and shrinking part of the market, based on information from
manufacturer interviews. Therefore, DOE is also considering accounting for the lowest part of
the market in the EISA qualified CSL as a simplifying alternative. Under this alternate
approach, there would be no difference between CSL 0 for Class A EPSs and CSL 0 for medical
and MADB EPSs in the engineering analysis. DOE is seeking stakeholder comment on which
approach to pursue or whether standard levels for medical EPSs should continue to be considered
in light of their critical lifesaving role.
ES.4.12 Scaling for EPS Representative Units and CSLs
DOE developed scaling relationships that it used both to analyze EPS representative unit
data as well as to scale CSLs from product class A1 to other product classes. In general, the
scaling methods for representative unit data were detailed whereas the scaling methods for CSLs
were simpler. For the representative unit data, DOE adopted the more detailed approach because
the detailed methods provide enhanced accuracy, which was paramount as it was the basis for
later analyses, including the LCC and NIA. For the CSL equations, DOE adopted the simpler
approach so that stakeholders could more easily evaluate the CSLs. DOE is seeking comment on
these approaches in general and whether to develop more detailed CSL equations in particular.
To analyze representative unit data, DOE developed and applied the following scaling
relationships:
Efficiency by nameplate output power
Efficiency by nameplate output voltage
Efficiency by output cord length
Cost by nameplate output power
Cost by output cord length
Markups to MSP
DOE did not develop scaling relationships related to no-load mode power consumption or
cost by production volume. DOE is seeking comment on the scaling relationships in general and
its decision on which scaling relationships to develop.

44

ES.4.13 Cost Estimates for Product Classes not Fully Analyzed


DOE developed estimates of incremental costs for BC product classes not fully analyzed
using the extrapolation methods described in chapter 5. For EPS product classes not fully
analyzed, DOE used the cost estimates it developed for the representative units as follows: for
product classes A2 and A4 DOE used engineering results for the 2.5 W representative unit,
and for product class A3 DOE used engineering results for the 18 W representative unit. The
underlying assumption is that the incremental costs associated with improving the efficiency of
products in a non-analyzed product class are equal to those associated with improving the
corresponding representative unit. DOE invites comment on these approaches.
ES.4.14 Market Data
DOE based its estimates of BC and EPS shipments on the estimated shipments of the
applications they power. DOE relied on published reports, comments from stakeholders, and its
own estimates to estimate shipments for each application. DOE also conducted a market survey
of over 50 common applications to determine the prevalence of BCs and EPSs and their
characteristics. DOE used these characteristics to estimate BC and EPS shipments for each
product class and representative unit. DOE seeks comments on its estimates of BC and EPS
shipments for each application and also the shipments in each product class and for each
representative unit.
ES.4.15 Product Lifetimes
DOE developed estimates of the average lifetime of each BC and EPS application
examined in the preliminary analysis. DOE welcomes comment on these estimates and on the
use of individual application lifetimes to generate LCC results. In particular, DOE seeks
comment on the accuracy of using the distribution of lifetimes from various applications as
opposed to sampling from a continuous distribution of lifetime values for a given representative
unit.
ES.4.16 Estimated Markups and Incremental Markup Assumptions
The confidential nature of the markups applied to BCs and EPSs necessitated that DOE
estimate markups using data from publicly available corporate financial filings. DOE seeks
comment on the accuracy of its estimated markups and specifically seeks comment on the
assumptions and methods used to calculate markups applied to baseline product costs (baseline
markups) and markups applied to cost increases resulting from standards (incremental markups).
ES.4.17 Sensitivity Analysis Methodology and Results
DOE conducted sensitivity analyses to quantify the relationship between certain critical
inputs to the life-cycle cost (LCC) analysis and the resultant weighted-average LCC savings for
BCs and EPSs. DOE conducted a similar analysis as part of the national impact analysis. DOE
seeks comment on the sensitivity analysis results and methodology.

45

ES.4.18 Use in the Commercial Sector


DOE recognizes that some BC and EPS applications under consideration in this
rulemaking, notably notebook computers, are also used extensively in the commercial sector.
The preliminary analysis does not differentiate between applications used in the commercial and
residential sectors. DOE intends to examine the extent to which applications are used in the
commercial sector during the NOPR stage of the analysis. DOE requests comments on the nature
of the commercial share of application shipments, and the BC and EPS shipments that are made
for those applications in the commercial sector. DOE also seeks comment on how usage profiles
differ between residential and commercial users. DOE also recognizes that many portable BC
applications, such as MP3 players and mobile phones, may be sold to residential users but
charged in the commercial sector and thus incur costs associated with commercial utility rates.
DOE proposes to ignore usage of residential products in the commercial sector in its energy cost
savings estimations. DOE seeks comments on this proposal.
ES.4.19 Prospects for Substitution in Response to Standards
DOE assumed that prescribing a standard for BCs and EPSs will not result in competing
technologies being substituted for them. Competing technologies include fuel cells, internal
power supplies (IPS), primary batteries, Universal Serial Bus (USB) systems, and others. DOE
believes that because application and device portability is important to consumers they will
generally not substitute competing technologies for BCs. DOE addresses the use of USB power
for BC applications, but requests comments on the degree to which an amended EPS standard
might cause USB power to be substituted for EPS power. DOE also seeks comment on its
assumption that substitution to other technologies will not occur in response to BC and EPS
standards.
ES.4.20 Usage Profiles
DOE developed usage profiles to help estimate the energy consumption of BCs and EPSs
and the energy savings potential of standards for those products. EPS usage profiles consist of
time spent in active and no-load mode. Because an application can demand different levels of
power, time in active mode was further divided by application state. Each application state was
associated with a loading point, expressed as a percentage of nameplate output power. In
chapter 7 of the preliminary TSD, DOE identified five loading points to apply to applications if
test data were not available. DOE seeks comments on the usage profiles and loading points used
for each EPS application. BC usage profiles consist of time spent in active, maintenance, nobattery, and off mode and the number of complete charge cycles completed over 24 hours.
Application-specific BC usage profiles were developed for each application. DOE welcomes
comment on all of these usage profiles.
ES.4.21 Selecting Battery Charger Usage Profiles for Use in Determining Compliance
with Standards
To create an integrated energy performance metric for BCs requires that DOE identify a
single usage profile for each product class. DOE seeks comment on how these usage profiles
should be determined.

46

ES.4.22 Shipment Trends


In forecasting future shipments of BCs and EPSs, DOE assumed that current per-capita
consumption of BCs and EPSs would remain constant. In other words, shipment growth is set
equal to expected population growth during the next 33 years. DOE seeks comment on this
assumption.
ES.4.23 Base Case Efficiency Distributions
To estimate what proportion of the BC and EPS markets would be affected by a standard
at a particular level, DOE developed a base case efficiency forecast. Based on a consideration of
several factors discussed in chapter 9, DOE forecasts modest improvement in the efficiency of
Class A EPSs between now and 2013 and no improvement thereafter (in the absence of amended
Federal standards). DOE forecasts no improvement in BC efficiency between now and 2013
because it identified no significant drivers of efficiency. DOE welcomes comment on this
forecast for use in calculating NES and NPV.
ES.4.24 Quantifying Non-Consumer Benefits
DOE acknowledges that there may be certain benefits resulting from the improved
efficiency of BCs and EPSs that are not captured directly by the products' users. These include
both environmental and energy security-related externalities that are not reflected in energy
prices, such as reduced emissions of greenhouse gases and reduced use of fossil fuels for
electricity generation. DOE invites comments on the nature and magnitude of such benefits.
ES.4.25 Impact on Manufacturers
DOE requests comment on its identification of key issues and requests data and
information from interested parties that can assist in evaluating the potential impact of standards
on manufacturers presented in chapter 12.
ES.4.26 Cumulative Regulatory Burdens
DOE recognizes and seeks to mitigate the overlapping effects on manufacturers of new or
amended DOE standards and other regulatory actions affecting the same equipment. DOE will
analyze and consider the impact on manufacturers of multiple, product-specific regulatory
actions in the NOPR. DOE invites comment on how regulations applicable to BCs and EPSs may
contribute to manufacturers cumulative regulatory burden. DOE also requests information on
the cost of low- and high-efficiency BCs and EPSs that are expected to be compliant with these
regulations so that DOE can assess how these requirements may affect payback calculations.
ES.4.27 Valuing Airborne Emission Reductions
DOE will conduct an environmental assessment as part of the next phase of the standards
rulemaking, the NOPR. The primary environmental effects of energy conservation standards for
BCs and EPSs are to reduce power plant emissions resulting from reduced consumption of
electricity. DOE will assess these environmental effects by using NEMS-BT to provide key
inputs to its analysis. The portion of the environmental assessment that will be produced by

47

NEMS-BT considers carbon dioxide (CO2), sulfur dioxide (SO2), nitrogen oxides (NOX), and
mercury (Hg). DOE also plans to monetize the emission reductions due to standards consistent
with methods used in recent DOE standards rulemakings pertaining to commercial clothes
washers and other consumer products and commercial equipment. For example, in the case of
CO2, the most recent interagency estimates of the potential global benefits resulting from
reduced CO2 emissions in 2010 were $4.7, $21.4, $35.1, and $64.9 per metric ton in 2007
dollars. DOE invites comment on possible approaches to monetizing the emission reductions
expected to result from energy conservation standards.

48

CHAPTER 1 INTRODUCTION

TABLE OF CONTENTS
1.1
1.2
1.3
1.4

Document Purpose .......................................................................................... 1-1


History of Battery Charger and External Power Supply Rulemakings ........... 1-1
Process for Setting Energy Conservation Standards ....................................... 1-3
Structure of the Document .............................................................................. 1-6
LIST OF TABLES

TABLE 1-1 ANALYSES BY RULEMAKING STAGE ............................................................ 1-5

1-i

CHAPTER 1 INTRODUCTION
1.1

DOCUMENT PURPOSE

This preliminary technical support document (TSD) is a stand-alone report that presents
the technical analyses that the U.S. Department of Energy (DOE) has conducted in preparation
for setting energy conservation standards for battery chargers (BCs) and external power supplies
(EPSs). The public is invited to comment on these analyses, either in writing or orally at a public
meeting on October 13, 2010. Details about the public meeting and instructions for submitting
written comments are contained in the notice of public meeting (NOPM) published in the
Federal Register on September 15, 2010. DOE will review the comments it receives and revise
and update these analyses prior to publishing a notice of proposed rulemaking (NOPR) in the
Federal Register.
1.2

HISTORY OF BATTERY CHARGER AND EXTERNAL POWER SUPPLY


RULEMAKINGS

Title III of the Energy Policy and Conservation Act (EPCA) (42 U.S.C. 6291, et seq.) sets
forth a variety of provisions designed to improve energy efficiency. Part A of Title III (42 U.S.C.
62916309) establishes the Energy Conservation Program for Consumer Products Other Than
Automobiles. The consumer products subject to this program (referred to as covered
products) include BCs and EPSs. Section 135 of the Energy Policy Act of 2005 (EPACT 2005),
Pub. L. 109-58, amended sections 321 and 325 of EPCA by inserting definitions for BCs and
EPSs and directing the Secretary of Energy to carry out three activities: (1) establish test
procedures, (2) hold a scoping workshop to discuss plans for developing energy conservation
standards, and (3) conduct a determination analysis for energy conservation standards for BCs
and EPSs. (42 U.S.C. 6295(u))
DOE complied with the first of these requirements by publishing the test procedure final
rule on December 8, 2006. 71 FR 71340. This rule included definitions and test procedures for
BCs and EPSs. DOE codified a test procedure for BCs in Title 10 of the Code of Federal
Regulations (CFR), Part 430, Subpart B, Appendix Y (Uniform Test Method for Measuring the
Energy Consumption of Battery Chargers) and a test procedure for EPSs in 10 CFR Part 430,
Subpart B, Appendix Z (Uniform Test Method for Measuring the Energy Consumption of
External Power Supplies).
Complying with the second requirement, DOE then published a NOPM and availability
of documentation for public review on December 29, 2006. 71 FR 78389. DOE made two
documents available on its website: Plans for Developing Energy Conservation Standards for
Battery Chargers and External Power Supplies and The Current and Future Market for Battery
Chargers and External Power Supplies. The public meeting, called a Scoping Workshop, was
held at DOEs Forrestal Building in Washington, DC, on January 24, 2007. As EPACT 2005
required, the workshop focused on DOEs plans for developing energy conservation standards
for BCs and EPSs. Information pertaining to the scoping workshop is available on DOEs
website at:
www.eere.energy.gov/buildings/appliance_standards/residential/battery_external_det_2006.html.

1-1

Regarding the third requirement, the President signed into law the Energy Independence
and Security Act of 2007 (EISA 2007), Pub. L. 110-140, on December 19, 2007, amending
sections 321, 323, and 325 of EPCA. These amendments required significant changes to the
determination analysis DOE had been conducting. Sections 301, 309, and 310 of EISA 2007
made several changes to EPCA related to BCs and EPSs.
Section 301 of EISA 2007 amended section 321 of EPCA by modifying definitions
concerning EPSs. EPACT 2005 had amended EPCA to define an EPS as an external power
supply circuit that is used to convert household electric current into DC current or lower-voltage
AC current to operate a consumer product.a (42 U.S.C. 6291(36)(A)) Section 301 of EISA 2007
further amended this definition by creating a subset of EPSs called Class A EPSs. EISA 2007
defined this subset as those EPSs that, in addition to meeting several other requirements common
to all EPSs, are able to convert to only 1 AC or DC output voltage at a time and have
nameplate output power that is less than or equal to 250 watts.b (42 U.S.C. 6291(36)(C)(i))
Section 301 also amended EPCA to establish minimum standards for Class A EPSs,
which became effective on July 1, 2008 (42 U.S.C. 6295(u)(3)(A)), and directed DOE to publish
a final rule by July 1, 2011, to determine whether to amend these standards. (42 U.S.C.
6295(u)(3)(D)) Section 301 further directed DOE to issue a final rule that prescribes energy
conservation standards for BCs or determine that no standard is technically feasible or
economically justified. (42 U.S.C. 6295(u)(1)(E)(i)(II))
In satisfaction of this requirement, DOE is bundling BCs and Class A EPSs together in a
single rulemaking proceeding to consider appropriate energy conservation standards for these
products. DOE published the Notice of Public Meeting and Availability of Framework
Document for Battery Chargers and External Power Supplies on June 4, 2009. 74 FR 26816.
DOE then held a public meeting to receive comment on the framework documentc on July 16,
2009 (hereafter referred to as the framework document public meeting). The present preliminary
analysis represents the next stage in the rulemaking process.
Section 309 of EISA 2007 further amended section 325(u)(1)(E) of EPCA, instructing
DOE to issue no later than two years after EISA 2007's enactment a final rule that determines
whether energy conservation standards shall be issued for external power supplies or classes of
external power supplies. (42 U.S.C. 6295(u)(1)(E)(i)(I)) However, as section 301 of EISA
simultaneously set standards for Class A EPSs, DOE interprets sections 301 and 309 jointly as a
requirement to determine, no later than December 19, 2009, whether additional energy
conservation standards shall be issued for EPSs that are outside the scope of the current Class A
standards, e.g., multiple-voltage EPSs. DOE determined that standards are warranted for nona

The terms AC and DC refer to the polarity (i.e., direction) and amplitude of current and voltage associated
with electrical power. For example, a household wall socket supplies alternating current (AC), which varies in
amplitude and reverses polarity. In contrast, a battery or solar cell supplies direct current (DC), which is constant in
both amplitude and polarity.
b
EISA 2007 defines a Class A EPS as an EPS that converts AC line voltage to only 1 lower AC or DC output, is
intended to be used with a separate end-use product, is in a different enclosure from the end-use product, is wired to
the end-use product, and has rated output power that is less than 250 watts. (42 U.S.C. 6291(36)(C)(i))
c
Energy Conservation Standards Rulemaking for Battery Chargers and External Power Supplies. May 2009.
Available at:
http://www1.eere.energy.gov/buildings/appliance_standards/residential/pdfs/bceps_frameworkdocument.pdf

1-2

Class A EPSs in a final rule published on May 14, 2010. 75 FR 27170. Standards for non-Class
A EPSs are thus being considered within the present rulemaking process.
Finally, section 310 of EISA 2007 amended section 325 of EPCA to establish definitions
for active mode, standby mode, and off mode. (42 U.S.C. 6295(gg)(1)(A)) This section also
directed DOE to amend its existing test procedures by December 31, 2008, to measure the
energy consumed in standby mode and off mode for both BCs and EPSs. (42 U.S.C.
6295(gg)(2)(B)(i)) Further, it authorized DOE to amend, by rule, any of the definitions for
active, standby, and off mode. (42 U.S.C. 6295(gg)(2)(A)) The Department presented its
amendments during a public meeting on September 12, 2008 (hereafter referred to as the standby
and off mode test procedure public meeting) and published them in the Test Procedures for
Battery Chargers and External Power Supplies (Standby Mode and Off Mode) Final Rule on
March 27, 2009. 74 FR 13318.
1.3

PROCESS FOR SETTING ENERGY CONSERVATION STANDARDS

Under EPCA, when DOE studies new or amended standards, it must consider, to the
greatest extent practicable, the following seven factors:
(1) the economic impact of the standard on the manufacturers and consumers of the
products subject to the standard;
(2) the savings in operating costs throughout the estimated average life of the products
compared to any increase in the prices, initial costs, or maintenance expenses for the
products that are likely to result from the imposition of the standard;
(3) the total projected amount of energy savings likely to result directly from the
imposition of the standard;
(4) any lessening of the utility or the performance of the covered products likely to result
from the imposition of the standard;
(5) the impact of any lessening of competition, as determined in writing by the Attorney
General, that is likely to result from the imposition of the standard;
(6) the need for national energy conservation; and
(7) other factors the secretary considers relevant. (42 U.S.C. 6295(o)(2)(B)(i))
Other statutory requirements are set forth in 42 U.S.C. 6295(o)(1)(2)(A), (2)(B)(ii)(iii),
and (3)(4).
DOE considers the participation of interested parties a very important part of the process
for setting energy conservation standards. Through formal public notifications (i.e., Federal
Register notices), DOE encourages the participation of all interested parties during the comment
period in each stage of the rulemaking. Beginning with the framework document and during

1-3

subsequent comment periods, interactions among interested parties provide a balanced


discussion of the information that is required for the standards rulemaking.
Before DOE determines whether to adopt a proposed energy conservation standard, it
must first solicit comments on the proposed standard. (42 U.S.C. 6313(a)(6)(B)(i)) Any new or
amended standard must be designed to achieve significant additional conservation of energy and
be technologically feasible and economically justified. (42 U.S.C. 6313(a)(6)(A)) To determine
whether economic justification exists, DOE must review comments on the proposal and
determine that the benefits of the proposed standard exceed its burdens to the greatest extent
practicable, weighing the seven factors listed above. (42 U.S.C. 6295 (o)(2)(B)(i))
After the publication of the framework document, the energy conservation standards
rulemaking process involves three additional public notices that DOE publishes in the Federal
Register. The first of the rulemaking notices is a NOPM, which is designed to publicly vet the
models and tools used in the preliminary rulemaking and to facilitate public participation before
the NOPR stage. The second notice is the NOPR, which presents a discussion of comments
received in response to the NOPM and the preliminary analyses and analytical tools; analyses of
the impacts of potential new or amended energy conservation standards on consumers,
manufacturers, and the Nation; DOEs weighting of these impacts; and the proposed energy
conservation standards for each product. The third notice is the final rule, which presents a
discussion of the comments received in response to the NOPR, the revised analyses, DOEs
weighting of these impacts, the amended energy conservation standards DOE is adopting for
each product, and the effective dates of the amended energy conservation standards.
DOE published the Notice of Public Meeting and Availability of the Framework
Document on June 4, 2009. 74 FR 26816. The framework document, Energy Conservation
Standards Rulemaking Framework Document for Battery Chargers and External Power Supplies,
describes the procedural and analytical approaches DOE anticipated using to evaluate the
establishment of new or amended energy conservation standards for these products. This
document is available at:
www.eere.energy.gov/buildings/appliance_standards/residential/pdfs/bceps_frameworkdocumen
t.pdf.
Subsequently, DOE held a public meeting on July 16, 2009 to discuss procedural and
analytical approaches to the rulemaking. In addition, DOE used the public meeting to inform and
facilitate the involvement of interested parties in the rulemaking process. The analytical
framework presented at the public meeting described the different analyses, such as the
engineering analysis and the consumer economic analyses (i.e., the life-cycle cost (LCC) and
payback period (PBP) analyses), the methods proposed for conducting them, and the
relationships among the various analyses.

1-4

Table 1-1 Analyses by Rulemaking Stage


Market and technology assessment
Screening analysis
Engineering analysis
Energy use characterization
Product price determination
Life-cycle cost and payback period analyses
Life-cycle cost subgroup analysis
Shipments analysis
National impact analysis
Preliminary manufacturer impact analysis
Manufacturer impact analysis
Utility impact analysis
Employment impact analysis
Environmental assessment
Regulatory impact analysis

Preliminary

NOPR

Final Rule

During the July 2009 public meeting, interested parties commented about numerous
issues relating to the analyses listed in Table 1-1. Comments from interested parties submitted
during the framework document comment period elaborated on the issues raised during the
public meeting. DOE attempted to address these issues as part of its preliminary analyses and
summarized the comments and DOEs responses in chapter 2 of the TSD.
As part of the information gathering and sharing process, DOE interviewed BC and EPS
manufacturers and those who make associated end-use consumer products. DOE selected
companies that represented production of all types of products, ranging from small to large
manufacturers. DOE had four objectives for these interviews: (1) solicit manufacturer feedback
on the draft inputs to the engineering analysis; (2) solicit feedback on topics related to the
preliminary manufacturer impact analysis; (3) provide an opportunity, early in the rulemaking
process, for manufacturers to express their concerns to DOE; and (4) foster cooperation between
manufacturers and DOE.
DOE incorporated the information gathered during these interviews into its engineering
analysis (chapter 5) and its preliminary manufacturer impact analysis (chapter 12). Following the
publication of the preliminary analyses and the associated public meeting, DOE intends to hold
additional meetings with manufacturers as part of the consultative process for the manufacturer
impact analysis conducted during the NOPR phase of the rulemaking.
DOE developed spreadsheets for the engineering, LCC, PBP, and national impact
analyses (NIA) for each product class, as well as spreadsheets for the market assessment, product
price determination, and energy use and end-use load characterization. The LCC workbook
calculates the LCC and PBP at various energy efficiency levels. The NIA workbook does the
same for national energy savings (NES) and national net present values (NPVs). All of these
spreadsheets are available on the DOE website for battery chargers and external power supplies:
http://www.eere.energy.gov/buildings/appliance_standards/residential/battery_external_std_2008

1-5

.html. DOE can also provide various quantitative outputs from its analyses in machine-readable
format upon request. For example, trial runs from the LCC Monte Carlo simulations could be
provided in such a format.
1.4

STRUCTURE OF THE DOCUMENT

The preliminary TSD describes the analytical approaches used in the preliminary analysis
and presents preliminary results. The TSD consists of 14 chapters, an executive summary, two
reports, and several appendices.
Executive Summary: Describes the rulemaking process, identifies the key
results of the preliminary analyses, and identifies the key issues for which
DOE seeks public comment that resulted from the preliminary analyses.
Chapter 1

Introduction: provides an overview of the appliance standards program


and how it applies to the battery charger and external power supply
rulemaking, and outlines the structure of the document.

Chapter 2

Analytical Framework: describes the methodology, the analytical tools,


and relationships among the various analyses, summarizes comments
DOE received on the framework document, and explains DOEs responses
to those comments.

Chapter 3

Market and Technology Assessment: characterizes the market for the


considered products and the technologies available for increasing energy
efficiency.

Chapter 4

Screening Analysis: identifies all the design options that improve


efficiency of the considered products, and determines which technology
options are viable for consideration in the engineering analysis.

Chapter 5

Engineering Analysis: discusses the methods used for developing the


relationship between increased manufacturer price and increased
efficiency.

Chapter 6

Product Price Determination: discusses the methods used for establishing


markups for converting manufacturer prices to final product prices.

Chapter 7

Energy Use Analysis: discusses the process used for generating energyuse estimates for the considered products as a function of standard levels.

Chapter 8

Life-Cycle Cost and Payback Period Analyses: discusses the effects of


standards on individual customers and users of the products and compares
the LCC and PBP of products with and without higher energy
conservation standards.

1-6

Chapter 9

Shipments Analysis: discusses the methods used for forecasting the total
number of products that would be affected by standards.

Chapter 10

National Impact Analysis: discusses the methods used for forecasting


national energy consumption and national consumer economic impacts in
the absence and presence of standards.

Chapter 11

Life-Cycle Cost Subgroup Analysis: discusses the effects of standards on


different subgroups of consumers and compares the LCC and PBP of
products with and without higher energy conservation standards for these
consumers.

Chapter 12

Preliminary Manufacturer Impact Analysis: discusses the effects of


standards on the finances and profitability of product manufacturers.

Chapter 13

Utility Impact Analysis: discusses the effects of standards on electric and


gas utilities.

Chapter 14

Employment Impact Analysis: discusses the effects of standards on


national employment.

Report

Environmental Assessment for Battery Chargers and External Power


Supplies: discusses the effects of standards on three pollutants sulfur
dioxide (SO2), nitrogen oxides (NOX), and mercury as well as carbon
emissions.

Report

Regulatory Impact Analysis for Battery Chargers and External Power


Supplies: discusses the impact of non-regulatory alternatives to efficiency
standards.

Appendices:
3A

Battery Charger and External Power Supply Applications

3B

Battery Charger and External Power Supply Efficiency Programs

3C

Evaluation Methods for Differentiating Wall Adapters with Charge


Control

5A

External Power Supply Test Data

5B

Battery Charger Test Data

5C

Manufacturer Interview Guides

1-7

7A

Battery Charger and External Power Supply Usage Profiles, Application


States, and Loading Points

7B
8A

Battery Charger and External Power Supply Application-Level Unit


Energy Consumption Values
User Instructions for LCC and PBP Spreadsheets

8B

Supplementary LCC and PBP Results

8C

End-Use Application Inputs for the LCC

8D

Residential Discount Rate Distributions

10A

Net Present Value under Alternative Electricity Price Scenarios

12A

MIA Manufacturer Interview Guide

15A

Interagency Report on the Social Cost of Carbon

1-8

CHAPTER 2. ANALYTICAL FRAMEWORK, COMMENTS FROM

INTERESTED PARTIES, AND DOE RESPONSES

TABLE OF CONTENTS
CHAPTER 2. ANALYTICAL FRAMEWORK, COMMENTS FROM INTERESTED

PARTIES, AND DOE RESPONSES .............................................................................. 2-1

2.1
Introduction.......................................................................................................... 2-1

2.1.1 Overview of Analyses.............................................................................. 2-1

2.1.2 Test Procedures........................................................................................ 2-3

2.2
Scope of Coverage ............................................................................................... 2-5

2.2.1 What Is an External Power Supply? ........................................................ 2-5

2.2.2 What Is a Battery Charger?...................................................................... 2-9

2.2.3 What Is a Consumer Product?................................................................ 2-10

2.3
Market and Technology Assessment ................................................................. 2-12

2.3.1 Market Assessment ................................................................................ 2-12

2.3.2 Product Classes ...................................................................................... 2-14

2.3.3 Technology Assessment......................................................................... 2-29

2.4
Screening Analysis............................................................................................. 2-34

2.4.1 Technology Options Screened Out ........................................................ 2-35

2.5
Engineering Analysis ......................................................................................... 2-36

2.5.1 Approaches to Conducting the Engineering Analysis ........................... 2-36

2.5.2 Engineering Analysis for Class A External Power Supplies ................. 2-37

2.5.3 Engineering Analysis for Non-Class A EPSs ........................................ 2-48

2.5.4 Engineering Analysis for Battery Chargers ........................................... 2-55

2.6
Markups to Determine Product Price................................................................. 2-63

2.7
Energy Use Analysis.......................................................................................... 2-63

2.7.1 External Power Supply Usage and Energy Consumption...................... 2-64

2.7.2 Battery Charger Usage and Energy Consumption ................................. 2-65

2.8
Life-Cycle Cost and Payback Period Analysis .................................................. 2-65

2.8.1 Approach................................................................................................ 2-66

2.8.2 Life-Cycle Cost Inputs........................................................................... 2-67

2.8.3 Payback Period....................................................................................... 2-72

2.8.4 Sensitivity Analysis ............................................................................... 2-73

2.9
Shipments Analysis............................................................................................ 2-73

2.9.1 Shipments Forecast ................................................................................ 2-74

2.9.2 Efficiency Forecast ................................................................................ 2-74

2.10 National Impact Analysis................................................................................... 2-76

2.10.1 Approach................................................................................................ 2-76

2.10.2 NIA Inputs ............................................................................................. 2-77

2.10.3 National Energy Savings Analysis......................................................... 2-77

2.10.4 Net Present Value Analysis ................................................................... 2-78

2.11 Life-Cycle Cost Subgroup Analysis .................................................................. 2-78

2.12 Preliminary Manufacturer Impact Analysis....................................................... 2-79

2.12.1 Sources of Information for the MIA ...................................................... 2-79

2.12.2 Industry Cash-Flow Analysis................................................................. 2-80

2-i

2.13
2.14
2.15

2.16
2.17

2.12.3 Manufacturer Subgroup Analysis .......................................................... 2-80

2.12.4 Competitive Impacts Assessment .......................................................... 2-81

2.12.5 Cumulative Regulatory Burden ............................................................. 2-81

2.12.6 Preliminary Results for the Manufacturer Impact Analysis .................. 2-81

2.12.7 Impacts on Shipments ............................................................................ 2-81

2.12.8 Timing.................................................................................................... 2-81

2.12.9 Efficiency Levels ................................................................................... 2-82

2.12.10 Regulatory Approval........................................................................... 2-82

2.12.11 Conversion Costs ................................................................................ 2-82

2.12.12 Efficiency as a Selling Point ............................................................... 2-82

2.12.13 Profit Margins ..................................................................................... 2-82

Utility Impact Analysis ...................................................................................... 2-82

Employment Impact Analysis............................................................................ 2-83

Environmental Assessment................................................................................ 2-83

2.15.1 Carbon Dioxide...................................................................................... 2-83

2.15.2 Sulfur Dioxide........................................................................................ 2-84

2.15.3 Nitrogen Oxides ..................................................................................... 2-84

2.15.4 Mercury.................................................................................................. 2-84

2.15.5 Particulate Matter................................................................................... 2-85

2.15.6 Monetization of Emissions Reduction Benefits..................................... 2-85

Regulatory Impact Analysis............................................................................... 2-86

Additional Comments from Interested Parties................................................... 2-87

2.17.1 Temporary Exemptions from Standards for Replacement Parts............ 2-87

2.17.2 Labeling Requirements and Marking Protocol ...................................... 2-87

2.17.3 Standards Compliance and Enforcement ............................................... 2-88

2.17.4 Publishing Public Comments on the Internet......................................... 2-88

2.17.5 Public Meeting Webcast ........................................................................ 2-88

LIST OF TABLES
TABLE 2.1 CLASS-SETTING CHARACTERISTICS CONSIDERED ................................. 2-15

TABLE 2.2 CLASS A EPS PRODUCT CLASSES.................................................................. 2-27

TABLE 2.3 MULTIPLE VOLTAGE EPS PRODUCT CLASSES .......................................... 2-27

TABLE 2.4 HIGH POWER EPS PRODUCT CLASSES ......................................................... 2-27

TABLE 2.5 MEDICAL EPS PRODUCT CLASSES................................................................ 2-27

TABLE 2.6 MADB EPS PRODUCT CLASSES ...................................................................... 2-27

TABLE 2.7 BC PRODUCT CLASSES USED IN THE PRELIMINARY ANALYSIS. ......... 2-29

TABLE 2.8 CLASS A EPS REPRESENTATIVE PRODUCT CLASS AND SCALED

PRODUCT CLASSES................................................................................................... 2-37

TABLE 2.9 CLASS A EPS REPRESENTATIVE UNITS ....................................................... 2-38

TABLE 2.10 SUMMARY OF CLASS A EPS CANDIDATE STANDARD LEVELS OF

EFFICIENCY ................................................................................................................ 2-38

TABLE 2.11 BASELINE (CSL 0) EISA 2007 ACTIVE-MODE EFFICIENCY FOR PRODUCT

CLASS A1 ..................................................................................................................... 2-39

2-ii

TABLE 2.12 EISA 2007 STANDARD EQUATION FOR MAXIMUM NO-LOAD MODE

POWER CONSUMPTION............................................................................................ 2-39

TABLE 2.13 ENERGY STAR 2.0 EQUATION FOR MINIMUM ACTIVE-MODE

EFFICIENCY FOR BASIC OUTPUT VOLTAGE EPSS ............................................ 2-39

TABLE 2.14 ENERGY STAR 2.0 STANDARD EQUATION FOR NO-LOAD MODE POWER

CONSUMPTION........................................................................................................... 2-40

TABLE 2.15 CSL 2 INTERMEDIATE ACTIVE-MODE EFFICIENCY FOR PRODUCT

CLASS A1 ..................................................................................................................... 2-40

TABLE 2.16 CSL 2 INTERMEDIATE NO-LOAD MODE POWER CONSUMPTION FOR

PRODUCT CLASS A1.................................................................................................. 2-40

TABLE 2.17 CSL 3 BEST IN MARKET ACTIVE-MODE EFFICIENCY FOR PRODUCT

CLASS A1 ..................................................................................................................... 2-41

TABLE 2.18 CSL 3 BEST IN MARKET NO-LOAD MODE POWER CONSUMPTION FOR

PRODUCT CLASS A1.................................................................................................. 2-41

TABLE 2.19 CSL 4 MAX TECH ACTIVE-MODE EFFICIENCY FOR PRODUCT CLASS A1

........................................................................................................................................ 2-42

TABLE 2.20 CSL 4 MAX TECH NO-LOAD MODE POWER CONSUMPTION FOR

PRODUCT CLASS A1.................................................................................................. 2-42

TABLE 2.21 ACTIVE-MODE EFFICIENCY CSLS FOR LOW-VOLTAGE EPSS (PRODUCT

CLASSES A2 AND A4)................................................................................................ 2-47

TABLE 2.22 NO-LOAD POWER CSLS FOR AC/AC EPSS (PRODUCT CLASSES A3 AND

A4) ................................................................................................................................. 2-47

TABLE 2.23 MULTIPLE VOLTAGE EPS REPRESENTATIVE PRODUCT CLASSES..... 2-48

TABLE 2.24 MULTIPLE-VOLTAGE EPS REPRESENTATIVE UNITS.............................. 2-48

TABLE 2.25 HIGH POWER EPS REPRESENTATIVE PRODUCT CLASSES ................... 2-49

TABLE 2.26 HIGH-POWER EPS REPRESENTATIVE UNITS ............................................ 2-49

TABLE 2.27 PROPOSED MEDICAL EPS REPRESENTATIVE PRODUCT CLASSES..... 2-49

TABLE 2.28 POSSIBLE MEDICAL EPS REPRESENTATIVE UNITS................................ 2-49

TABLE 2.29 PROPOSED MADB EPS REPRESENTATIVE PRODUCT CLASSES ........... 2-50

TABLE 2.30 POSSIBLE MADB EPS REPRESENTATIVE UNITS ...................................... 2-50

TABLE 2.31 MULTIPLE-VOLTAGE EPS CANDIDATE STANDARD LEVELS OF

EFFICIENCY ................................................................................................................ 2-51

TABLE 2.32 HIGH-POWER EPS CANDIDATE STANDARD LEVELS OF EFFICIENCY2-51

TABLE 2.33 MEDICAL EPS CANDIDATE STANDARD LEVELS OF EFFICIENCY ...... 2-52

TABLE 2.34 MADB EPS CANDIDATE STANDARD LEVELS OF EFFICIENCY............. 2-53

TABLE 2.35 BC REPRESENTATIVE PRODUCT CLASSES ............................................... 2-56

TABLE 2.36 THE BC REPRESENTATIVE UNITS FOR EACH REPRESENTATIVE

PRODUCT CLASS ....................................................................................................... 2-57

TABLE 2.37 CANDIDATE STANDARD LEVELS FOR THE BC REPRESENTATIVE UNITS

........................................................................................................................................ 2-59

TABLE 2.38 THE BC REPRESENTATIVE UNITS FOR EACH REPRESENTATIVE AND

SCALED PRODUCT CLASS....................................................................................... 2-62

TABLE 2.39 CANDIDATE STANDARD LEVELS FOR THE SCALED BC

REPRESENTATIVE UNITS. ....................................................................................... 2-62

TABLE 2.40 SUMMARY OF INPUTS USED IN THE LIFE-CYCLE COST ANALYSIS .. 2-67

TABLE 2.41 MARKET-WEIGHTED LIFETIMES FOR REPRESENTATIVE UNITS........ 2-70

2-iii

TABLE 2.42 INPUT VARIATIONS FOR THE SENSITIVITY ANALYSIS......................... 2-73

TABLE 2.43 SUMMARY OF INPUTS USED IN THE NATIONAL IMPACT ANALYSIS.......

.................................................................................................................................................... 2-77

LIST OF FIGURES

FIGURE 2.1 FULL COST OF PRODUCT: BREAKDOWN OF PRODUCTION AND NON


PRODUCTION COSTS ................................................................................................ 2-43

2-iv

CHAPTER 2. ANALYTICAL FRAMEWORK, COMMENTS FROM

INTERESTED PARTIES, AND DOE RESPONSES

2.1

INTRODUCTION

The Energy Policy and Conservation Act of 1975, as amended (EPCA) requires the U.S.
Department of Energy (DOE) to promulgate energy conservation standards for consumer
products that achieve the maximum improvement in energy efficiency that the Department
determines is both technologically feasible and economically justified. 42 U.S.C.
6295(o)(2)(A). This chapter provides a description of the general analytical framework that DOE
is using to develop such standards for battery chargers (BCs) and external power supplies
(EPSs). The analytical framework is a description of the methodology, the analytical tools, and
relationships among the various analyses that are part of this rulemaking.
DOE developed this analytical framework and documented its findings in the Energy
Conservation Standards Rulemaking Framework Document for BCs and Class A EPSs (June 4,
2009). DOE announced the availability of the framework document in a Notice of Public
Meeting and Availability of a Framework Document published in the Federal Register on June 4,
2009. 74 FR 26816. DOE presented the analytical approach to interested parties during a public
meeting held on July 16, 2009. In addition to describing the analytical framework, this chapter
summarizes written and oral comments DOE received from interested parties on the framework
document and explains DOEs responses to those comments. The framework document is
available at
www1.eere.energy.gov/buildings/appliance_standards/residential/battery_external_std_2008.htm
l
DOE published a Notice of Proposed Determination (NOPD) for non-Class A EPSs on
November 3, 2009, followed by a final determination for non-Class A EPSs that was published
on May 14, 2010. This final determination, which was issued as a rule consistent with the
particular requirements set out in EPCA, indicated that energy conservation standards for nonClass A EPSs appear to be technologically feasible and economically justified, and are likely to
result in significant energy savings. 74 FR 56928 and 75 FR 27170, respectively. Consequently,
the Department is including non-Class A EPSs in the energy conservation standards for BCs and
EPSs. In its determination, DOE also indicated that it would address certain comments raised by
interested parties in the forthcoming preliminary analysis since they raised general issues related
to BCEPS.
Significant changes in analytical approach from what was described in the framework
document and determination analysis are also noted in this chapter. Lastly, note that the
executive summary of the preliminary TSD contains a list of issues for which DOE seeks public
comment. Those issues are explained in detail in the relevant sections below.
2.1.1 Overview of Analyses
The analyses performed as part of this rulemaking stage and reported in this preliminary
technical support document (preliminary TSD) are listed below.

2-1

A market and technology assessment to characterize the relevant product markets and
existing technology options, including prototype designs.
A screening analysis to review each technology option and determine if it: is
technologically feasible; is practical to manufacture, install, and service; would adversely
affect product utility or product availability; or would have adverse impacts on health and
safety.
An engineering analysis to develop cost-efficiency relationships that show the
manufacturers cost of achieving increased efficiency. DOE determines the increased cost
to the consumer through an analysis of engineering markups, which convert manufacturer
production cost (MPC) for EPSs and efficiency-related manufacturing cost (ERMC) for
BCs to manufacturer selling price (MSP). DOE uses different price structures for EPSs
and BCs because EPSs are contained within a single enclosure that has a single MSP
whereas BCs often share an enclosure with other non-BC related components. Thus,
DOE uses ERMC to address the prices of components pertinent to BCs only.
A markup analysis to develop distribution channel markups that relate the MSP to the
cost to the consumer.
An energy use analysis to determine the annual energy use in the field of the considered
products.
Life-cycle cost (LCC) and payback period (PBP) analyses to calculate, at the consumer
level, the discounted savings in operating costs throughout the estimated average life of
the covered products, compared to any increase in the installed cost for the products
likely to result directly from imposition of the standard. A sensitivity analysis to
determine LCC results in a high-savings case and a low-savings case.
A shipments analysis to forecast product shipments, which are then used to calculate the
national impacts of standards on energy, net present value (NPV), and future
manufacturer cash flows.
A national impact analysis (NIA) to assess the aggregate impacts at the national level of
the NPV of total consumer LCC and national energy savings (NES).
A preliminary manufacturer impact analysis to assess the potential impacts of energy
conservation standards on manufacturers, such as impacts on capital conversion
expenditures, marketing costs, shipments, and research and development costs.
The analyses DOE will perform in the subsequent Notice of Proposed Rulemaking
(NOPR) stage include those listed below. In addition, DOE will revise the analyses it performed
in the preliminary stage based on comments and new information received.
An LCC subgroup analysis to evaluate variations in customer characteristics that might
cause a standard to affect particular consumer sub-populations, such as low-income
households, differently than the overall population.
A manufacturer impact analysis to estimate the financial impact of standards on
manufacturers and to calculate impacts on competition, employment, and manufacturing
capacity.
A utility impact analysis to estimate the effects of proposed standards on electric, gas, or
oil utilities.
An employment impact analysis to assess the aggregate impacts on national employment.

2-2

An environmental impact analysis to provide estimates of the effects of amended energy


conservation standards on three pollutantssulfur dioxide (SO2), nitrogen oxides (NOX),
and mercury (Hg)as well as carbon emissions.
A regulatory impact analysis to present major alternatives to proposed amended energy
conservation standards that could achieve substantially the same regulatory goal at a
lower cost.
Note that non-Class A EPSs (discussed in section 2.2.1.1) are included in the preliminary
market and technology assessment, markup analysis, and shipments analysis, but are not
included in the preliminary engineering analysis or in those analyses that make use of
engineering outputs, i.e., the LCC and NIA. DOE intends to conduct a comprehensive analysis of
potential standards for these products in the NOPR stage of the rulemaking.
2.1.2 Test Procedures
Section 323 of EPCA (42 U.S.C. 6293) sets forth generally applicable criteria and
procedures for DOEs adoption and amendment of test procedures, which manufacturers of
covered products must use to certify to the DOE that their products comply with EPCA energy
conservation standards and to quantify the efficiency of their products. Also, these test
procedures must be used whenever testing is required in an enforcement action to determine
whether covered products comply with EPCA standards.
DOE has adopted test procedures for both BCs and EPSs. These are described in the
sections below. Further discussion of testing and efficiency metrics and their application to the
analysis of achievable performance can be found in sections 2.3.3.1 for EPSs and 2.3.3.3 for
BCs.
2.1.2.1 EPS Test Procedures
On December 8, 2006, DOE codified a test procedure final rule for EPSs in appendix Z to
subpart B of 10 CFR Part 430 (Uniform Test Method for Measuring the Energy Consumption
of External Power Supplies). 71 FR 71340, 71368. DOEs test procedure, based on the
ENERGY STAR EPS test procedure, measures active-mode efficiency and no-load mode
(standby mode) power consumption. In the standby and off mode test procedure NOPR for BCs
and EPSs, 73 FR 48054, 48063 (August 15, 2008), DOE proposed to amend the EPS test
procedure to add a measurement of power consumption in off mode, where, if the EPS has an
on-off switch, the EPS is connected only to mains and the switch is turned off. The NOPR also
included a proposed test procedure for multiple-voltage EPSs. The amendments for standby and
off mode were included in the final rule, published March 27, 2009. 74 FR 13318, 13335. In the
final rule DOE also reserved a section of the test procedure for multiple-voltage EPSs. DOE is
considering amending the EPS test procedure as part of its revision to the BC test procedure
NOPR published on April 2, 2010. 75 FR 16958.
Active mode conversion efficiency is the ratio of output power to input power. DOE
averages the efficiency at four loading conditions25, 50, 75, and 100 percent of maximum
rated output currentto assess the performance of a single-voltage EPS. For multiple-voltage
EPSs, the test procedure NOPR proposed that the test procedure yield the efficiency

2-3

measurement at the same four loading conditions without averaging them. 75 FR 16958, 16974.
For single-voltage EPSs, DOE measures the power consumption of the EPS when disconnected
from the consumer product, which is termed no-load power consumption. If the EPS has an onoff switch, the switch is on when conducting the measurement. DOE proposed to treat no-load
and off modes similarly for multiple-voltage EPSs in the NOPR.
2.1.2.2 BC Test Procedures
On December 8, 2006, DOE adopted a test method to measure the efficiency of battery
chargers. 71 FR 71340, 71368. This test method, based on the U.S. Environmental Protection
Agencys (EPA) ENERGY STAR Test Methodology for Determining the Energy Performance
of Battery Charging Systems, measures the power consumed by BCs in maintenance and nobattery, as well as the energy recovered from the battery during discharge, by calculating an
energy ratio.
In the December 8, 2006, Test Procedure Final Rule, DOE stated that it intended to study
further BC active mode energy consumption and reserved a section in the test procedure (section
4(b) of appendix Y to subpart B of 10 CFR Part 430). 71 FR 71340, 71360. During the
framework document public meeting, DOE presented several potential approaches to measuring
battery charger energy consumption in active mode. Numerous interested parties submitted
comments on active mode measurement in general but noted several specific issues:
Harmonization of DOEs BC active mode test procedure with the BC test procedure
adopted by the California Energy Commission (CEC)
Limiting testing to U.S. line-voltage AC input (115 volts at 60 hertz)
Shortening the charge and maintenance mode testing period to less than 24 hours
Incorporating usage profiles into the test procedure
Measuring charger output energy instead of battery output energy
Measurements at alternative depths-of-discharge to better reflect consumer use
Test procedure rulemaking schedule
Appropriate voltage for testing multiple-voltage BCs
Power factor measurement
DOE took the comments on the above topics under advisement as it developed
amendments to its existing BC test procedure to accommodate energy consumption in active
mode. Comments relevant to the proposal, as well as the proposed amendments themselves, are
discussed in depth in the BC active mode test procedure NOPR, published on April 2, 2010. 75
FR 16958, 16963-76.

2-4

If adopted, these proposed amendments to the BC test procedure would discontinue use
of the non-active energy ratio in favor of metrics related to energy consumption in each of the
energy-consuming modes of operation of a BC. As described above, these include active,
maintenance, standby, and off modes; thus, the test procedure would return four separate values.
How these four mode-specific metrics would be combined for the purpose of an energy
conservation standard is discussed further in section 2.3.3.3.
2.2

SCOPE OF COVERAGE
This standards rulemaking concerns all BCs and EPSs, which are defined as follows:
The term battery charger means a device that charges batteries for consumer
products, including battery chargers embedded in other consumer products. (42
U.S.C. 6291(32))
The term external power supply means an external power supply circuit that is used
to convert household electric current into DC current or lower-voltage AC current to
operate a consumer product. (42 U.S.C. 6291(36)(A))

DOE has received numerous comments on scope in connection with this rulemaking and
the related test procedure and determination rulemakings. For example, Vivitar requested clear
interpretations of statutory definitions to ease uncertainty among manufacturers. (Vivatar, Pub.
Mtg. Tr., No. 14 at pp. 115, 375) PG&E commented that there is a tradeoff between simplicity
and maximum energy savings. (PG&E, Pub. Mtg. Tr., No. 14 at p. 118) AHAM similarly
commented on the impossibility of covering every possible EPS or BC product and encouraged
DOE to maintain simplicity when deciding issues of scope. (AHAM, Pub. Mtg. Tr., No. 14 at
p. 95)
Ambiguities inherent in the statutory definitions have left interested parties with
questions about which products are EPSs and which are BCs. The following subsections are
intended to answer these questions.
2.2.1 What Is an External Power Supply?
As noted above, an EPS is an external power supply circuit that is used to convert
household electric current into DC current or lower-voltage AC current to operate a consumer
product. Thus, DOE understands an EPS to have four key attributes:
1. An EPS is outside (external to) the product it operates,
2. An EPS attaches to and obtains power from mains (household electric power),
3. An EPS performs power conversion but does not contain additional charge control
functions, and
4. An EPS powers a consumer product (discussed in section 2.2.3 below).

2-5

Of these four attributes, the third has generated the most discussion because of the
difficulty of clearly establishing the presence of charge control. DOE recognizes that some wall
adaptersa do more than simply convert household electric current into DC current or lowervoltage AC current. They also perform charge control functionsb as part of a battery charging
system. As proposed in the framework document, DOE has decided to exclude these wall
adapters from the universe of EPSs with the understanding that their energy performance will be
addressed by BC standards.
2.2.1.1 Class A Versus Non-Class A EPSs
EPCA provides definitions for EPS and Class A EPS. Non-Class A EPSs, then, are those
devices that fit the definition of an EPS but do not fit the definition of a Class A EPS.
In general, a Class A EPS is a device that:
1.
2.
3.
4.
5.
6.

is designed to convert line voltage AC input into lower voltage AC or DC output;


is able to convert to only one AC or DC output voltage at a time;
is sold with, or intended to be used with, a separate end-use product that
constitutes the primary load;
is contained in a separate physical enclosure from the end-use product;
is connected to the end-use product via a removable or hard-wired male/female
electrical connection, cable, cord, or other wiring; and
has nameplate output power that is less than or equal to 250 watts.

However, the term Class A EPS does not include any device that:
I.

II.

requires Federal Food and Drug Administration listing and approval as a medical
device in accordance with section 513 of the Federal Food, Drug, and Cosmetic
Act (21 U.S.C. 360c); or
powers the charger of a detachable battery pack or charges the battery of a
product that is fully or primarily motor operated. (42 U.S.C. 6291(36)(C))

Using this framework, DOE has identified four types of non-Class A EPSs:
1. Multiple-Voltage EPSs: These devices are able to convert to more than one AC or DC
output voltage at a time.
2. High-Power EPSs: These devices have a nameplate output power greater than 250
watts.
3. Medical EPSs: These devices are used to power medical devices regulated by the
Food and Drug Administration.
4. MADB EPSs: These devices provide power to the battery chargers of motorized
applications and detachable battery (MADB) packs.
a

The term wall adapter refers to the device between mains and an application, that may or may not fit the
statutory definition of external power supply in that it may, in some cases, provide charge control in addition to
power conversion.
b
Charge control functions include current and temperature sensing, timing, and current limiting, all of which are
included in battery charging systems for safety reasons.

2-6

Medical EPSs include EPSs that power medical devices and EPSs that are themselves
medical devices. A literal reading of EPCA would exclude from Class A only those EPSs that
are themselves medical devices. The term class A external power supply excludes any device
that requires Federal Food and Drug Administration listing and approval as a medical device in
accordance with section 513 of the Federal Food, Drug, and Cosmetic Act (21 U.S.C. 360c). 42
U.S.C. 6291(36)(C)(ii)(I) However, a search of FDAs product classification database for
power supply reveals only one EPS that is a medical deviceauxiliary power supply
(alternating current (AC) or direct current (DC)) for external transcutaneous cardiac pacemakers.
Furthermore, all EPSs used with medical devices must meet the special requirements of UL
60601 (Underwriters Laboratories standard for power supplies for medical devices).
Accordingly, because the exclusion applies to any device covered by the FDA's listing and
approval requirements, DOE interprets EPCA to also exclude from Class A those EPSs that
power medical devices.
MADB EPSs provide power to the chargers of motorized applications and/or detachable
battery packs. A detachable battery is one that is contained in a separate enclosure from the
product and intended to be removed or disconnected from the product for recharging. See 42
U.S.C. 6291(52). MADB EPSs are employed by certain household appliances such as handheld
vacuums, personal care products such as shavers, and DIY power tools. MADB EPSs were
referred to as EPSs for BCs in the determination analysis of non-Class A EPSs. DOE now uses
the term MADB to highlight the two unique aspects that EPCA specifies: motorized
applications (MA) and detachable batteries (DB).
DOE received two comments on what should be considered a detachable battery. Vivitar
commented that batteries that are either physically or electrically disconnected for recharging
should be considered detachable. (Vivitar, Pub. Mtg. Tr., No. 14 at p. 80) PG&E commented that
the term should include AA batteries for flashlights. (PG&E, Pub. Mtg. Tr., No. 14 at p. 88)
Vivitars comment regarded camera batteries recharged inside the camera by a wall
adapter. Because the battery is electrically disconnected from the functional part of the camera
during charging, and a user cannot operate the camera while the battery is charging, the battery
can be considered detachable. The wall adapter could therefore be considered a MADB nonClass A EPS, pending the results of other DOE tests seeking to determine the presence of charge
control, described in section 2.2.1.3.
The flashlight batteries PG&E referred to are not detachable for purposes of the present
discussion. Universal rechargeable batteries (AA, AAA, etc.) that are removed from the product
they power and inserted into a separate universal BC for charging may be detachable, but are
not battery packs. Thus, the EPS that powers such a universal BC is not excluded from Class A.
2.2.1.2 Wall Adapters Without Charge Control
Wall adapters without charge control are EPSs, even when they are part of a battery
charging system. This is true of both Class A and non-Class A EPSs. This decision is consistent
with Approach A, for which DOE received widespread support from interested parties. Under
this approach, described in detail in the framework document, only wall adapters without charge

2-7

control would be subject to EPS standards. Methods for identifying the presence of charge
control in wall adapters are discussed in section 2.2.1.3 and chapter 3 of the TSD.
FRIWO, Motorola, and PG&E were generally unified in their support for Approach A.
Commenters agreed that it is appropriate to subject wall adapters without charge control to EPS
standards because they can be used for other applications besides battery charging. (FRIWO, No.
21 at p. 1; PG&E, Pub. Mtg. Tr., No. 14 at p. 62; Motorola, No. 25 at p. 1; PG&E, No. 20 at p. 3)
PG&E commented that a significant percentage of wall adapters used with battery chargers lack
charge control and thus would be considered EPSs under Approach A, including those for
cellular telephones, cordless telephones, notebook computers, portable media players, and other
digital electronics. (PG&E, No. 20 at p. 3) PG&E added that due to EPS efficiency standards in
California, which covered all wall adapters without regard for the presence of charge control,
nearly all wall adapters considered in the framework document should already be compliant to
EISA 2007 levels (Roman number IV) at least in California, which is a significant portion of the
U.S. economy. (PG&E, No. 20 at p. 6)
AHAM and Wahl Clipper asked DOE to consider an alternative approach for wall
adapters that are used with motorized applications and/or detachable batteries. Specifically, these
commenters requested that DOE not consider these devices EPSs at all, irrespective of the
presence of charge control. (AHAM, No. 16 at p. 2-3; Wahl Clipper, No. 23 at p. 1) Under this
approach there would be no MADB EPSs. PTI implied in a comment submitted in response to
the NOPD that it would also support such an approach when it stated its preference for an
approach that excludes from coverage under EPS standards any wall adapter that is incorporated
as part of a BC. (PTI, EERE-2009-BT-DET-0005 No. 5 at pp. 1-2) As noted in the determination
published on May 14, 2010, under the approach recommended by AHAM and Wahl Clipper,
wall adapters that power the chargers of detachable battery packs or charge the batteries of
products that are fully or primarily motor-operated would not be subject to EPS standards while
those wall adapters (without charge control) that power other battery-charged applications would
be subject to Class A EPS standards. 75 FR 27170. DOE is concerned that this would create two
distinct definitions of an EPS that would prevent one from readily identifying a particular wall
adapter as being an EPS until it is known whether it powers the charger of a detachable battery
pack or charges the battery of a product that is fully or primarily motor operated. In the absence
of information that would alleviate this concern, DOE continues to consider wall adapters
without charge control to be EPSs, whether they are Class A or non-Class A.
2.2.1.3 Identification of Charge Control in Wall Adapters
In the framework document, DOE sought stakeholder comment on four possible criteria
for identifying charge control in a wall adapter: short-circuit operation, voltage regulation, noload voltage, and no-battery operation. DOE received comments on these criteria and alternative
methods for identifying charge control.
Commenters challenged the criteria that DOE proposed to identify charge control by
claiming that the criteria presented by DOE are impractical. PG&E commented that the electrical
tests proposed by DOE for identifying a wall adapter with charge control function are ineffective
and recommended that DOE rely on physical characteristics to determine which wall adapters
should be excluded from treatment as EPSs. PG&E noted that these physical characteristics

2-8

could include actual charge control circuitry or a state-of-charge indicator. (PG&E, No. 20 at
pp. 3-4; PG&E, Pub. Mtg. Tr., No. 14 at p. 61) In response to the notice of proposed
determination, PG&E also identified a chemistry-type selector switch as a physical characteristic
the presence of which could be used to determine whether charge control is present in a wall
adapter. (PG&E, EERE-2009-BT-DET-0005 No. 7 at p. 7) Similarly, FRIWO commented that
DOE should consider the presence of LED charge indicator lights and temperature sensors.
(FRIWO, No. 21 at p. 1) Finally, PG&E commented that it is possible to have no charge control
at all in the battery charging systemneither in the wall adapter nor in subsequent enclosures.
(PG&E, Pub. Mtg. Tr., No. 14 at p. 130)
Although some battery charging systems may not have components explicitly devoted to
charge control, all require this function, and DOE intends to define it. Specifically, DOE is
considering adopting the EPA criteria for establishing the presence of charge control in a wall
adapter. DOE would consider a wall adapter with charge control to have batteries or battery
packs that physically attach directly or a battery chemistry or type selector switch AND an
indicator light or state of charge meter. These criteria are modeled after those contained in the
BC standards proposal developed by PG&E.1 Nonetheless, DOE is also evaluating other criteria
that might be more appropriate to determine whether charge control is present in a wall adapter.
Details of these other criteria can be found in chapter 3 of the preliminary TSD and appendix 3C.
2.2.2 What Is a Battery Charger?
As noted above, a BC is a device that charges batteries for consumer products, including
battery chargers embedded in other consumer products. Like EPSs, BCs are used with
consumer products. However, unlike EPSs, the statutory definition of BCs also includes products
not powered from mains. Further, BCs may be wholly embedded in another consumer product,
wholly separate from another consumer product, or partially inside and partially outside another
consumer product.
2.2.2.1 Battery Chargers with Wall Adapters
When a wall adapter is present, it is considered part of the BC. This approach is
consistent with the statutory definition of a BC and has received broad support from interested
parties. Including the wall adapter in the BC is a key part of Approach A, which, as indicated
above, received support from a number of stakeholders. Further, AHAM and PG&E commented
that DOE should not separate components of a battery charging system when testing its
efficiency. (AHAM, No. 16 at p. 1; PG&E, Pub. Mtg. Tr., No. 14 at p. 60) PG&E commented
that the efficiency of a wall adapter is not indicative of the energy consumption of the entire
battery charging system. (PG&E, No. 20 at p. 13)
DOE received two comments requesting alternative approaches. Vivitar requested that
DOE set separate standards for wall adapters for battery charging applications, rather than
including them within the scope of either BC or EPS standards. Delta-Q commented that DOE's
scope of analysis should be limited to power conversion functions (e.g., only the wall adapter)
and should not include charge control. (Vivitar, Pub. Mtg. Tr., No. 14 at p. 112; Delta-Q, No. 15
at p. 1)

2-9

After examining these alternative approaches proposed by interested parties, DOE


determined that they would be inconsistent with the statutory framework set by Congress.
Regarding Vivitars request, DOE is considering new and amended standards for BCs and EPSs.
Regarding Delta-Qs request, DOE cannot limit the scope of its BC analysis only to power
conversion functions because a BCs power-consuming functions extend beyond power
conversion to include charge control. Often, charge control is included not in the wall adapter,
but in the end-use consumer products, and the statute clearly specifies that DOE is to include
within its scope BCs embedded in other consumer products.
2.2.2.2 DC-Powered Battery Charging Systems
While many BCs are powered from AC mains, others have DC inputs, including those
that obtain power from a computers universal serial bus (USB) port and those that connect to an
automobiles cigarette lighter receptacle.
DOE received several comments from interested parties regarding DC-powered BCs.
FRIWO asked that DC-powered battery chargers be included within the scope of the BC
rulemaking (FRIWO, No. 21 at p. 1), whereas AHAM disagreed, asking that DOE exclude
cigarette-lighter and USB adapters. (AHAM, No. 16 at p. 5) PG&E. commented that wall
adapters sold as an after-market accessory for a USB BC should be considered part of a BC
when sold by the same manufacturer. (PG&E, No. 20 at p. 3) Motorola commented that mobile
phones are often charged from DC sources such as computers. (Motorola, Pub. Mtg. Tr., No. 14
at p. 99)
DOE is including all DC-powered BCs within the scope of this rulemaking. The BC
definition, unlike the EPS definition, does not specify that the input power to the device must be
AC. Many battery-charged consumer electronics can obtain power from a USB port, which
provides power at 5 V DC (nominal). USB power sources include notebook and desktop
computers, USB hubs, standalone USB wall adapters, and even the chairs in some airport
lounges. Because USB power sources are becoming more varied and widespread, DOE is
focusing its BC analysis on that which is between the USB port and the battery, without regard
for the USB ports power source. Standalone wall adapters with USB outputs are EPSs and, as
such, are addressed as part of the EPS standards analysis.
2.2.3 What Is a Consumer Product?
An EPS operates a consumer product, and a BC charges batteries for consumer products.
What, then, is a consumer product? The statutory definition of a consumer product is any article
other than an automobile, as defined in section 32901(a)(3) of title 49, that consumes energy or
water and which, to any significant extent, is distributed in commerce for personal use or
consumption by individuals. (42 U.S.C. 6291(1))
The definition of automobile cited above states that except as provided in 49 U.S.C.
32908, the term automobile means a 4-wheeled vehicle that is propelled by fuel, or by
alternative fuel, manufactured primarily for use on public streets, roads, and highways and rated
at less than 10,000 pounds gross vehicle weight, except (A) a vehicle operated only on a rail line;
(B) a vehicle manufactured in different stages by 2 or more manufacturers, if no intermediate or

2-10

final-stage manufacturer of that vehicle manufactures more than 10,000 multi-stage vehicles per
year; or (C) a work truck. (49 U.S.C. 32901(a)(3)) The term alternative fuel includes
electricity. (49 U.S.C. 32901(a)(1)(J)) Given this definition, DOE will not examine BCs for
electric automobiles, nor will it examine those for low-speed vehicles, which are similar to
golf cars but are manufactured with additional safety features for on-road use. These vehicles are
covered as automobiles under Federal Motor Vehicle Standard No. 500: Low-Speed Vehicles. 49
CFR 571.500.
DOE received a number of comments from interested parties related to the definition of
consumer product and the scope of BC and EPS analyses and standards. Delta-Q asked whether
high-power battery chargers would be covered by BC standards, (Delta-Q, No. 15 at p. 1) while
Wahl requested that DOE exempt BCs with output power greater than 250 watts until an
appropriate test procedure is developed. Wahl further requested that DOE exclude medical
device BCs. (Wahl, No. 23 at p. 1) Motorola asked whether industrial products such as
emergency/first-responder radios would be covered by standards, (Motorola, Pub. Mtg. Tr., No.
14 at p. 349) and Schumacher requested that DOE include within its scope BCs used for
automotive and other conveyance applications. (Schumacher, No. 27 at p. 2)
High-power BCs and BCs for medical devices are indeed within the scope of this
standards rulemaking, provided that the BCs charge batteries for consumer products as defined
above. The BCs used to charge golf car batteries, for example, are covered products because golf
cars are distributed in commerce for personal use or consumption by individuals and are not
automobiles, as defined above; that is, they are not manufactured primarily for use on public
streets, roads, and highways. DOE recognizes that many golf cars are in fact sold to businesses
for use in golf course fleets. However, those sold to businesses are indistinguishable from those
sold to individuals and, thus, all of the associated BCs are included within DOEs scope. BCs for
emergency/first-responder radios, in contrast, are not covered products because emergency/first
responder radios and other products intended only for commercial, industrial, and institutional
use are not consumer products.
ASAP and PG&E recommended that DOE incorporate a wide variety of BCs in its
standards analysis. (ASAP, Pub. Mtg. Tr., No. 14 at p. 39; PG&E, Pub. Mtg. Tr., No. 14 at p.
49, 57) PG&E further recommended that DOE include in its analysis a variety of end-use
applications, charger designs, and battery energy capacities and all common battery chemistries.
PG&E provided a list of at least 30 consumer charger types in 14 different product categories for
DOE to consider. (PG&E, Pub. Mtg. Tr., No. 14 at p. 57; PG&E, No. 20 at p. 8) Finally,
Motorola commented that there are currently no provisions for testing the efficiency of systems
with backup batteries for data retention and indicated it feels a wide range of products should
be evaluated (Motorola, No. 25 at p. 2) AHAM warned DOE not to stray from the consumer
products definition in NAECA. (AHAM, Pub. Mtg. Tr., No. 14 at p. 352)
In its preliminary analysis, DOE characterized a variety of BCs and EPSs for a wide
range of consumer products, including many that it had not examined previously. DOE expanded
the scope of its analysis to cover over 50 applications for BCs and over 70 for EPSs. DOE tested
and examined many charger designs ranging from inductively-coupled chargers to golf car
chargers. DOE also examined the most common battery chemistries including lithium ion, nickel
metal hydride, and sealed lead acid. DOE characterized BCs for batteries with energy capacities

2-11

ranging from 0.5 watt-hours to over 3 kilowatt-hours. While DOE analyzed uninterruptible
power supplies that provide emergency backup power, it did not examine systems with backup
batteries for data retention such as those embedded in desktop computers.
2.3

MARKET AND TECHNOLOGY ASSESSMENT

2.3.1 Market Assessment


DOE developed a qualitative and quantitative characterization of BC and EPS markets
and industry structure. DOE based this market assessment, presented in full detail in chapter 3,
on publicly available information, including purchased market research reports, as well as data
and information submitted by interested parties. Where specific data was not available, DOE
developed estimates based on industry studies, and data from similar products or associated
applications. Issues addressed include market characteristics and trends, existing regulatory and
non-regulatory efficiency improvement initiatives both in the United States and abroad, product
distribution, manufacturer market share, and other manufacturer and industry characteristics.
BCs and EPSs are typically sold to consumers as components of other end-use products
(applications). DOE identified more than 80 BC and EPS applications and focused the market
assessment on the markets for these products for two reasons:
The demand for applications drives the demand for BCs and EPSs, because BCs and
EPSs are nearly always integrated into, bundled with, or otherwise intended to be
used with a given application.
Because most BCs and EPSs are not stand-alone products, their usage profiles, energy
consumption, and power requirements are all determined by the associated
application.
After identifying the products that use BCs and EPSs, DOE collected data on product
lifetimes, BC/EPS characteristics, and the number of units shipped annually for sale in the
United States for these applications. DOE then translated these application-level data to the
shipments, lifetimes, and performance characteristics of BCs and EPSs themselves.
2.3.1.1 Bar Code Scanners, Mobile Computers, and Wireless Headphones
CEA submitted comments on the NOPD regarding standards for non-Class A EPSs in
which it asked DOE to examine the possibility that EPSs for bar code scanners, mobile
computers, and wireless headphones are MADB EPSs. (CEA, EERE-2009-BT-DET-0005 No. 9
at p. 2) Upon investigation it appears that bar code scanners and mobile computers with batteries
intended to be removed for charging are used in the commercial sector and are not consumer
products. Batteries for wireless headphones do not appear to fit the definition of detachable
battery and, therefore, EPSs used with these products are in Class A. Both the BCs and EPSs
used with wireless headphones are included in the preliminary analysis.

2-12

2.3.1.2 Shipments and Lifetimes Estimates


DOE received several comments related to the product lifetime and shipment estimates
contained in the draft technical report that accompanied the framework document. PG&E
commented that DOE, significantly underestimated the 2007 sales of EPSs and consumer
battery chargers in the U.S. (PG&E, No. 20 at p. 11) In a follow-up comment, CEC provided
DOE with U.S. shipments for several product applications and provided estimates for BC and
EPS shipments and installed bases. (CEC et al., No. 19 at p. 1) Responding to these comments,
DOE has expanded the scope of BCs covered and examined more EPS applications since the
draft technical report was published. DOE also received additional data from PG&E and
incorporated it into its analysis. DOE estimates that a total of 402 million BCs and 301 million
EPSs were shipped in 2008 for final sale in the United States.
DOE received two comments in response to the NOPD on the prevalence of high power
EPSs. PG&E and CEC both commented that the number of high power EPSs (those with
nameplate output power greater than 250 watts) is likely to increase in the future as applications
such as game consoles, fast chargers, and other home electronics demand increasing amounts of
power. (PG&E, EERE-2009-BT-DET-0005 No. 7 at p. 2; CEC et al., EERE-2009-BT-DET-0005
No. 8 at p. 1) To date, DOE has identified only one application for high power EPSs, amateur
radio equipment, but has noted the existence of an LED monitor that uses an EPS with nameplate
output power of exactly 250 watts. DOE will continue to monitor the market for high power
EPSs and update its analysis in the NOPR stage.
BCs and EPSs are designed to outlast their respective applications, and because users
replace the BC or EPS when they replace an application, DOE assumes BC/EPS lifetimes are
typically equivalent to the lifetimes of their applications. DOE received support for this
assumption from interested parties. (PG&E, No. 20 at p. 10) One notable exception to this
assumption may apply to mobile phones. Several mobile phone manufacturers have agreed to
adopt the GSMA Universal Charging Solution by 2012, which will allow the same EPS to be
used with multiple phones. This voluntary standard is anticipated to eliminate the need to
package an EPS with each mobile phone and would result in a longer lifetime for those EPSs.
More information about the GSMA standard can be found in section 3.2.3 of the TSD.
2.3.1.3 Properly Representing BCs and EPSs Paths to Market
Given the wide array of applications that use BCs and EPSs, the paths to market are
varied and complex, but typically consist of: (1) component manufacturers, (2) BC/EPS
manufacturers (sometimes referred to as original device manufacturers or ODMs), (3) end-use
product manufacturers (sometimes referred to as original equipment manufacturers or OEMs),
and (4) end-use product retailers/distributors. Not all BCs or EPSs are distributed in the same
manner. Some follow a less direct path to market, such as those distributed by electrical
component distributors. Others follow a more direct path to market , such as those BCs
manufactured in-house by OEMs, which distribute them along with an application.
DOE presented its models for the distribution channels of BCs and EPSs at the
framework public meeting in July 2009 and received useful stakeholder comments. AHAM
explained that the distribution channels for BCs and EPSs are complicated but similar in

2-13

structure. Differences arise in what share of the market takes which distribution path. Further,
there are manufacturers of BC applications that sell products both to other manufacturers as well
as directly to consumers. (AHAM, Pub. Mtg. Tr., No. 14 at p. 231) PTI explained that many enduse product manufacturers (OEMs) do not manufacture the EPS or BC portions of the products
that they sell, although they may maintain engineering control over the BCs and EPSs. (PTI,
Pub. Mtg. Tr., No. 14 at p. 240) DOE recognizes the similarities between BC and EPS paths to
market as well as the complexity of these paths and made changes to its model to better reflect
the nature of the BC and EPS markets and industries.
2.3.1.4 Updates to BC and EPS Efficiency Distributions
In the draft technical report that accompanied the framework document, DOE presented
its understanding of the efficiency of BCs and EPSs on the market prior to 2008. PG&E
commented that DOE should update its efficiency distributions to represent the current state of
the market (PG&E, No. 20 at p. 12) and provided DOE with data from BC tests it conducted. To
characterize the distribution of BC and EPS shipments by efficiency level in 2009, DOE
examined data provided by PG&E and from tests DOE performed (see chapter 5). DOE then
weighted the test data by application shipments to arrive at BC and EPS efficiency distributions.
The efficiency distributions DOE derived from the bases for the base case efficiency forecasts
discussed in chapter 9.
2.3.2 Product Classes
When necessary, DOE divides covered products into classes by the type of energy used,
the capacity of the product, and any other performance-related feature that justifies different
standard levels, such as features affecting consumer utility. (42 U.S.C. 6295(q)) DOE then
conducts its analysis and considers establishing or amending standards to provide separate
standard levels for each product class.
In the framework document, DOE summarized the various product classes used by EISA
2007, ENERGY STAR, and the draft analysis prepared by DOE for the 2007 determination on
BCs and EPSs. DOE invited comment from interested parties on the product classes presented
and also sought suggestions regarding other possible product class-setting criteria that should be
used in establishing BC and EPS energy conservation standards.
Although the comments of interested parties concerned only BCs, the criteria proposed
are in some cases also applicable to EPSs. Therefore, the discussion of the characteristics is
organized by whether the characteristics can be used to set product classes for both BCs and
EPSs, BCs only, or EPSs only. These characteristics are listed in Table 2.1.

2-14

Table 2.1 Class-Setting Characteristics Considered


Characteristics Not
Characteristics
Considered Due to
Common to EPSs
Statutory
Characteristics
and BCs
Limitations
Unique to EPSs
Output power
Usage
Type of power
conversion
Output voltage
Topology
Medical use
Output cord
Price
length
Used with battery
chargers of
motorized
applications and
detachable
batteries

Characteristics
Unique to BCs
Battery voltage
Battery capacity
Battery energy
Charge rate
Battery chemistry
Inductive coupling
Cord-cordless
operation
Additional
functionality

2.3.2.1 Potential Product Class Criteria Applicable to Both BCs and EPSs
In general, PG&E. commented that DOE should develop a standard that is a continuous
function (as opposed to discrete steps in standard levels), similar to the EISA 2007 standards for
EPSs. According to PG&E, standards based on a continuous function would remove the
incentive to redesign a product that falls near a product class boundary, such that it falls in a
different product class and is subject to a more lenient standard. (PG&E, No. 20 at p. 7)
DOE acknowledges this potential for circumvention at the boundary between two product
classes. Therefore, DOE has sought to minimize the opportunities for circumvention by (1)
attempting to set the smallest number of product classes while complying with the requirements
of 42 U.S.C. 6295(q), and (2) proposing standard levels thatwhere feasibleare continuous
across product class boundaries.
Maximum Output Power. The primary function of both BCs and EPSs is providing
power; therefore, the main capacity-related characteristic that could be used for setting product
classes is maximum output power. This parameter determines the sizes of the power-handling
components of a power converter, and ultimately the size of the converter itself.
For EPSs, maximum output power is listed on the nameplate. EPSs with greater
nameplate output power offer the consumer greater capacity and tend to have higher active mode
average efficiency. EPSs with lower nameplate output power tend to have lower no-load power
consumption.
DOE acknowledges that nameplate output power significantly affects utility and
efficiency and therefore different nameplate output powers may warrant unique product classes.
However, rather than create distinct product classes by nameplate output power, DOE proposes
to follow the precedent set by EISA 2007 and ENERGY STAR 2.0 and intends to set an
efficiency standard level that is a continuous function of nameplate output power for Class A
EPSs.

2-15

One of the conditions by which EPCA defines a Class A EPS is that it has nameplate
output power that is less than or equal to 250 watts, hence DOE considers high power EPSs
with nameplate output power greater than 250 watts to be non-class A EPSs.
(42 U.S.C. 6291(36)(C)(i)(VI)) The ability to provide power at high wattages is unique to high
power EPSs. Currently, the most common use of high power EPSs is to power amateur radio
equipment. Similar to Class A EPSs, DOE intends to set a standard level that is a continuous
function of nameplate output power for high-power EPSs.
For BCs, maximum output power limits how quickly a BC can charge a given battery as
described by the following inequality:

PMAX

C BATT VBATT

Eq. 2.1

Where:

PMAX = the maximum output power, in watts,


CBATT = the battery capacity, in ampere-hours,
VBATT = the battery voltage, in volts, and
r
= the charge rate, in units of C or 1/hour.
AHAM, PTI, and Delta-Q commented that output power should be used as a criterion for
setting product classes. (AHAM, No. 16 at p. 3; Delta-Q, No. 15 at p. 1; PTI, Pub. Mtg. Tr., No.
14 at p. 287) AHAM and Delta-Q elaborated by specifying that DOE should place BCs with
output power greater than 250 watts (AHAM, No. 16 at p. 3) and 1000 watts (Delta-Q, No. 15 at
p. 1) in separate product classes.
As mentioned above, output power is a measure of BC capacity; furthermore, BCs with
different output powers may not be able to meet the same standard in maintenance and nobattery modes. Because output power is a capacity-related characteristic that tends to affect
efficiency, it meets the requirements of 42 U.S.C 6295(q), and DOE considered it in the
development of its BC product classes, presented in chapter 3.
Output Voltage. Many consumer products cannot function on power from mains, which
is at 115 volts; instead, they rely on power converters to convert mains power to a lower voltage.
In general, power converters with a higher output voltage will be more efficient. Many of the
losses that occur in a power converter are functions of current since the current increases as
voltage decreases for a given output power, as shown in: power (P) = current (I) voltage (V).
For EPSs, the output voltage is listed on the device nameplate, and is determined by the
characteristics of the application. For instance, certain applications such as monitors and printers
are designed to draw power from an EPS. Their components, such as the LCD screen and print
engines, have specific voltage requirements that the EPS helps to satisfy by providing power at
the required voltage.
ENERGY STAR 2.0 acknowledged that EPSs with lower output voltage are inherently
less efficient by setting a less stringent level of active mode average efficiency for EPSs with low
voltage and high current outputs. ENERGY STAR 2.0 defined low voltage models as EPSs

2-16

with nameplate output voltage less than 6 volts and nameplate output current greater than or
equal to 550 milliamps. ENERGY STAR publishes test results for qualifying EPSs that confirm
that many low-voltage EPSs would not have qualified at the standard-voltage level.2 In the
preliminary analysis, DOE set product classes for EPSs with standard voltage output and low
voltage output. DOE proposes to use the ENERGY STAR criteria to define low voltage EPSs
because the criteria adequately capture many low-voltage EPSs in the market.
Nameplate output voltage does not affect EPS no-load mode power consumption because
there is no output current in no-load mode. Therefore, DOE did not consider the effect of output
voltage on no-load mode power consumption in setting product classes.
Most EPSs can only output one voltage at a time, which is a requirement to be considered
Class A EPSs; however, some multiple-voltage EPSs output power at two or more voltages
simultaneously. Applications often contain different types of circuits that require different
voltages. OEMs choose whether to create different voltages within the application or to purchase
an EPS that provides multiple voltages. Typically, it is more cost-effective to create different
voltages within the application; however, in some instances manufacturers prefer to use multiplevoltage EPSs. DOE intends to create a separate product class for non-Class A multiple-voltage
EPSs because of their unique utility.
For BCs, the power converter output voltage is determined by the voltage of the battery,
as the BC must output a high enough voltage to fully charge the battery. However, because the
motors or electronics of the associated end-use product is powered by the rechargeable battery
have been designed given a particular battery voltage, changing the voltage could severely
impact consumer utility, forcing the redesign of a product and substitution of motors and
electronics for perhaps less desirable ones.
Numerous commenters, including Delta-Q, Wahl, FRIWO, and PTI recommended
classifying BCs by voltage. (Delta-Q, No. 15 at p. 1; Wahl, No. 23 at p. 2; FRIWO, No. 21 at
p. 2; PTI, Pub. Mtg. Tr., No. 14 at p. 287) FRIWO further specified divisions between the
voltage-based product classes based on the cell voltage of the currently used rechargeable
battery chemistry, namely lithium-ion and lithium-polymer. According to FRIWO, batteries
with voltages at 3.7, 7.4, 11.1, 18.5, 25.9, and greater than 25.9 volts should each be placed in a
separate product class. (FRIWO, No. 21 at p. 2) On the other hand, PG&E noted that if a BC
standard allows low voltage products to be less efficient, it will encourage design of less
efficient, low voltage products, and vice-versa. (PG&E, No. 20 at p. 7; PG&E, Pub. Mtg. Tr.,
No. 14 at p. 302)
Although PG&E is correct in noting that voltage-based product classes may result in less
stringent standards and potentially lower the efficiency for low-voltage BCs, such a result may
be required to ensure consumer utility. In particular, while a stringent standard on (less efficient)
low-voltage BCs may lead to substitution of higher-voltage BCs for lower-voltage BCs and
higher BC energy savings, it would also require the redesign of the end-use applications served
by those BCs. Those end-use applications lie outside of the current scope of this rulemaking.
Furthermore, in certain applications, the internal voltage of the product will not change
regardless of the battery voltage. In a notebook computer, for example, the processor voltage

2-17

must be 1.1 volts. Increasing the BC voltage would only increase the difference between the
battery voltage and the final processor voltage, which would increase the step-down conversion
losses inside the computer. Increasing the output voltage would, therefore, decrease the BC
losses, but may increase the losses in the system as a whole.
Because battery voltage impacts both the efficiency and utility of a BC, DOE will
consider it as a criterion for setting product classes. Furthermore, DOE agrees with FRIWO that
any divisions between voltage-based product classes should take into account the voltages of the
most common batteries, though whether these divisions should fall every 3.7 volts or more or
less frequently (e.g., every 1.2 volts to accommodate nickel-based chemistries) is discussed
further in section 2.3.2.3.
Output Cord Length. The length of the output cord affects both utility and efficiency:
longer output cords offer consumers more utility, so that they have more reach between the wall
adapter and the application. However, increasing cord length increases power loss and lowers
efficiency. In its comments, Wahl noted the length and stranding of the cord has an impact on the
BC efficiency (Wahl, No. 23 at p. 1), and should therefore be considered as a possible criterion
for setting product classes.
The output cord of an EPS can have an appreciable impact on its measured efficiency due
to resistive losses in the conductors. For example, DOE tests indicate that an output cord can
cause a 120-watt EPSs average efficiency to drop 0.8 percentage points, up to a maximum of
1.97 percentage points. This amount is significant in the EPS analysis, where a few percentage
points separate the most and least efficient units in the market at a given output power level.
The losses in the cord, and hence the efficiency of the EPS, are also affected by its crosssectional area. Specifically, the resistance of the output cord is proportionally related to the
inverse of its cross-sectional area: R 1 / A (where R is resistance, is the proportionality
symbol, and A is area). Although a thicker cord may be heavier, bulkier, and less flexible in
theory, no significant variations have been encountered in practice. DOE does not expect the
small variations to affect consumer utility.
DOE accounted for differences in output cord length through scaling when analyzing
EPS test data, teardown data and manufacturer data. However, DOE did not account for output
cord length when setting CSL efficiency equations for EPSs because the wide variety of output
cord lengths render it impractical to do that. DOE proposes not to include output cord length as a
factor in determining CSL equations.
In the BC analysis, output cord length and thickness were not taken into account because
even a 1.97 percentage point decrease in efficiency stemming from output cord losses is not
significant when compared to the active-mode efficiency, which can vary by 30 percentage
points or more across units of a single product class, and the annual energy consumption
(including energy consumption in all modes), which can vary by a factor of more than10.
Usage, Topology, and Price. Unlike some of the other products regulated by DOE
through energy conservation standards, BCs and EPSs are rarely the end-use application. This
fact is reflected, in part, by the statutory definition for BCs, which includes battery chargers

2-18

embedded in other consumer products in the BC definition. (42 U.S.C. 6291(32)) Similarly
the EPS definition specifies that an EPS is used . . . to operate a consumer product. (42 U.S.C.
6291(36)) As a result, the type of application that the BC or EPS serves, and in particular how
that application is used may also differentially impact the energy consumption of BCs and EPSs
in addition to the capacity- and utility-related characteristics described above. Both Wahl and
AHAM emphasized this fact by urging DOE to consider usage patterns when setting product
classes. AHAM also requested that DOE only consider minimal no-battery and maintenance
time when analyzing infrequently-used BCs. (Wahl, No. 23 at p. 2; AHAM, No. 16 at p. 4)
During the framework document public meeting, interested parties also commented on
the role BC topology should play in the preliminary analysis. Topology is a term used to describe
the general architecture of the power conversion stage of a BC. The choice of topology is
affected by considerations such as output voltage, output power, reliability, and even the
experience of the designer. While the topologies do not constrain the efficiency of a power
converter and its associated BC, they influence the choice of technology options for increasing
efficiency and affect the cost of efficiency improvement from the baseline. Because of the wide
variety of applications and limited time for analysis, PG&E recommended that DOE focus on
analyzing unique topologies and associated technology options instead. (PG&E, Pub. Mtg. Tr.,
No. 14 at p. 58) PTI, however, commented that since topology is transparent to consumers and
consumers do not buy products based on topology, it may not be a good criterion to consider for
product classes. (PTI, Pub. Mtg. Tr., No. 14 at p. 286)
Finally, Wahl commented that DOE should also consider the retail price of a BC when
creating product classes. (Wahl, No. 23 at p. 2)
Unlike the characteristics discussed previously, usage, topology, and price do not meet
the statutory criteria for setting separate standard levels for product classes under 42 U.S.C.
6295(q). In particular, although these characteristics may differentially impact the efficiency of
BCs and EPSs, they do not impact their capacity or utility; instead, price and topology are
impacted by the capacity- and utility-related features of a BC or EPS. By setting product classes
based directly on those features, as suggested above, DOE will also be taking into account the
effects of topology and price, while complying with the requirements of 42 U.S.C. 6291(q).
Finally, usage is not a characteristic of a BC or an EPS, but rather the end-use application
it serves. Therefore, even if a BC or EPS consumes more energy due to pairing with a heavily
used application, nothing prevents it from being paired with a lightly used application, and viceversa, as long as the two power converters have the same utility.
2.3.2.2 Potential Product Class Criteria Applicable Only to EPSs
Type of Power Conversion (AC/AC versus AC/DC). All EPSs receive input power from
mains in the form of alternating current (AC) and provide output power in the form of either AC
power or direct current (DC) power. This is another key functionality of an EPS, along with
providing power at a specific voltage. Again, the type of power conversion the EPS provides is
governed by its application. Applications such as mobile phones and laptops require DC power
to match the type of power provided by the battery. Applications that tend to use EPSs that

2-19

provide AC power, such as cordless phones, often have conversion circuitry within the
application for AC-DC conversion.
The type of power conversion is indicative of an EPSs internal circuitry, therefore its
capability to conserve energy. EPSs that provide AC output power typically contain only a
transformer. The no-load power losses of those EPSs are dominated by transformer core losses.
Alternatively, EPSs that provide DC power output typically contain a transformer as well as
overhead circuitry that control the flow of power through the EPS. Overhead circuitry provides
EPSs that function using DC-output power with the ability to reduce power consumption in noload mode in contrast to AC-output EPSs that typically cannot. Technically, it is possible to add
overhead circuitry to an EPS with AC output power to reduce EPS power consumption in noload mode; however, the overhead power consumption of that circuit would most probably
outweigh potential energy savings from the rest of the EPS. ENERGY STAR 2.0 acknowledges
this relationship by setting a less stringent level of no-load mode power consumption for EPSs
with AC output power. Although DOE did not receive comments from interested parties related
to EPS conversion type, DOE set different product classes for EPSs with AC output power and
DC output power in the preliminary analysis.
Use with Medical Equipment. Another condition by which EPCA defines a Class A EPS
is that it does not include any device that requires Federal Food and Drug Administration listing
and approval as a medical device in accordance with section 513 of the Federal Food, Drug, and
Cosmetic Act (21 U.S.C. 360c). (42 U.S.C. 6291(36)(C)(ii)(I)) Thus, all EPSs used with
medical devices must meet the special requirements of UL standard 60601, such that they are
approved by the Federal Food and Drug Administration (FDA)c. Use with medical devices is a
utility that is unique to medical EPSs, which are exempted from EISA 2007 efficiency standards.
Consequently, there may be medical EPSs in the market with efficiencies lower than required by
EISA 2007 for Class A EPSs. In the non-Class A EPS determination analysis, DOE tested two
medical EPSs as well as four Class A EPSs that DOE tentatively identified as being
electronically equivalent to medical EPSs. Of those six EPSs, only one of the Class A EPSs had
an efficiency level lower than that required by EISA 2007. DOE is considering testing more
medical EPSs for the NOPR analysis to further characterize the distribution of efficiency levels
in the medical EPS market.
Use with Battery Chargers of Motorized Applications and Detachable Batteries. EPCA
also excludes from the Class A EPS definition any device that powers the charger of a detachable
battery pack or charges the battery of a product that is fully or primarily motor operated.
(42 U.S.C. 6291(36)(C)(ii)(II)) By extension, these devices are exempted from having to meet
the efficiency standards for Class A EPSs set by EISA 2007. Examples of products that are fully
or primarily motor operated include portable rechargeable household appliances such as
handheld vacuums, personal care products such as shavers, and power tools. DOE considers
EPSs for battery chargers of motorized applications and detachable batteries" (MADB EPSs) to
have a unique utility from Class A EPSs because EISA 2007 excluded them from Class A EPS
coverage. 42 U.S.C. 6291(36)(C)(ii). Because there are no required efficiency levels for these
c

Underwriters Laboratories standard for power supplies for medical devices. More information on UL 60601 and
how it applies to medical EPSs is presented in section B. Scope of the Notice of Proposed Determination (NOPD)
for non-Class A EPSs published on May 14, 2010. 74 FR 56928.

2-20

types of EPSs, DOE believes at this time that MADB EPSs have efficiencies lower than those
established for Class A EPSs. In the non-Class A EPS determination analysis, DOE tested 12
Class A EPSs that it believed were electronically equivalent to MADB EPSs. Of those 12 EPSs,
only two of the Class A EPSs had efficiency levels lower than that required by EISA 2007. DOE
is considering testing MADB EPSs for the NOPR to further characterize the distribution of
efficiency levels in the MADB EPS market.
2.3.2.3 Potential Product Class Criteria Applicable Only to BCs
Battery Capacity. Similar to battery voltage, battery capacity impacts utility as well as
BC output power, the primary capacity-related characteristic of the BC. A battery with half the
capacity of another will power the same consumer product half as long. Wahl and PTI both
recommended that battery capacity could be used for setting product classes. (PTI, Pub. Mtg. Tr.,
No. 14 at p. 287; Wahl, No. 23 at p. 1) However, Wahl also commented that any variations in
efficiency due to battery capacity could also be addressed in the test procedure. (Wahl, No. 23 at
p. 1) Although the commenter did not specify how, one possible method could involve
normalizing the energy consumption result by the battery capacity.
Neither the current nor proposed DOE test procedure for battery chargers normalizes the
energy consumption of the battery charger by battery capacity. Therefore, any capacity-related
differences in energy consumption would have to be addressed through different standard levels
for each capacity-related product class. Because battery capacity impacts both the efficiency of
the BC and its utility, it can be used to establish product classes, and is discussed further in
chapter 3.
Battery Energy. Battery energy is the product of battery capacity and battery voltage as
described below:

E BATT

C BATT VBATT

Eq. 2.2

Where:

EBATT = the battery energy in watt-hours,


CBATT = the battery capacity, in ampere-hours, and
VBATT = the battery voltage, in volts.
Battery energy combines the effects of both battery capacity and voltage; PG&E.
suggested that this characteristic is most indicative of a battery charging system's actual utility.
(PG&E, No. 20 at p. 7) PTI joined PG&E. in recommending that battery energy be used to set
product classes. (PTI, Pub. Mtg. Tr., No. 14 at p. 287)
As with both battery voltage and battery capacity, battery energy is an important battery
charger characteristic that impacts not only BC efficiency but also BC utility and capacity (e.g., a
charger redesigned to charge a higher-energy battery in the same amount of time would see an
increase in output power). It merits further consideration because, as noted by PG&E, it
determines the amount of work that an end-use consumer product can perform, regardless of the
specifics of BC or application design. As described later in chapter 3, DOE used battery energy

2-21

as a proxy for the other potential product-class-setting characteristics, such as maximum output
power and battery capacity.
Charge Rate. Charge rate is equal to the charge current (in amperes) normalized by the
battery capacity (in ampere-hours), and reflects the speed at which a battery is being charged.
Under ideal conditions (i.e., ignoring cell inefficiency and cell equalization or other additional
charging steps), the charge time would be the inverse of the charge rate, which is expressed in
units of C or 1/hours.
Comments from interested parties diverged on whether to set product classes based on
charge rate. PG&E argued against a separate product class for slow (i.e., low-charge-rate)
chargers out of concern that a separate class would permit less efficient BCs despite the
availability of the technology improvement options for fast chargers that DOE listed in the
framework document could also be cost-effectively implemented in slow chargers. (PG&E, No.
20 at pp. 6 and 8) Wahl also commented that charge rate should not be used to create product
classes, despite its important role in BC design. (Wahl, No. 23 at pp. 12) Conversely, PTI and
AHAM spoke in favor of product classes based on charge rate, because it is more costly to
increase the efficiency of slow chargers than that of fast chargers. (AHAM, Pub. Mtg. Tr., No.
14 at pp. 309; PTI, Pub. Mtg. Tr., No. 14 287)
Charge rate affects the output power of the charger. For a given battery voltage and
capacity, a faster charger has a higher charge rate, and therefore a higher output power. This
relationship corresponds to a higher capacity for the purposes of this discussion, as the charger
can transfer more energy to the battery per unit time.
However, as mentioned above, output power also depends on the voltage, capacity,
and/or energy of the battery. Furthermore, the variation in output power due to the battery
parameters is much greater than that due to charge rate, as charge rates rarely fall outside the
range of 0.1 C to 1 C, while battery energies can span over five orders of magnitude from less
than 1 watt-hour to over 10,000 watt-hours. Because of its more limited effect on output power,
it is more accurate to consider charge rate as a utility characteristic. A BC with a higher charge
rate will charge the battery in less time, providing greater utility to the user.
Fast chargers, which have higher capacities and provide greater consumer utility, also
tend to be more efficient than slow chargers. Fast chargers typically use switched-mode power
supplies that have lower transformer core losses than slow chargers incorporating line-frequency
transformers followed by a rectifier. Additionally, slow chargers generally do not have
monitoring and cutoff circuitry. The lack of this circuitry means that once a battery is fully
charged, slow chargers tend to deliver excess current to the battery, which causes maintenance
mode power consumption to increase and further decreases efficiency.
Furthermore, with respect to the comments regarding technology optionsi.e., whether
technology options exist to make slow chargers as efficient as fast chargersthe information
available to DOE indicate that slow chargers do not provide any appreciable benefits when
compared to fast chargers. This observation, when coupled with the fact that slow chargers can
be readily replaced with fast chargers for the same applications, tends to support the view that
there is insufficient justification to support the creation of a separate product class based on

2-22

charge rate. As the name suggests, slow chargers tend to take longer to charge the same battery
as a fast charger and thus provide less utility to the consumer. DOE notes that replacement costs
are factored in its engineering analysis described later in chapter 5. Accordingly, in developing
its product class discussion found in Chapter 3, DOE did not consider classes based on charge
rates. Finally, because slow chargers do not provide any unique utility to consumers relative to
fast chargers, DOE cannot set less stringent standards for slow chargers under 42 U.S.C.
6295(q).
Battery Chemistry. Battery chemistry is another term for the battery compositioni.e.,
the materials that serve as the electrodes and electrolyte of the electrochemical cell. The
properties of these materials determine the cell voltage, cell efficiency, self-discharge rate, and
other properties of the battery. They also affect the design of the BC, including the detection of
charge completion, the need for termination (i.e., significantly decreasing the charging current
following full charge), and the extent of safety circuitry. Despite the impacts of battery chemistry
on the design of the battery charging system, Wahl commented that battery chemistry should not
be used to create product classes. (Wahl, No. 23 at p. 3) PTI disagreed, stating that lithium-based
chemistries should be differentiated from older chemistries. (PTI, Pub. Mtg. Tr., No. 14 at
p. 287)
Battery chemistries differ in, for example, energy density and maximum discharge
current, which can result in differences in consumer utility such as the length of time that a
battery can be stored between recharges, the weight of the battery, or the maximum mechanical
power output from a mechanical appliance, respectively. Nonetheless, the actual impact on users
of consumer products is minor. For example, power tools, shavers, and electric toothbrushes can
use both lithium- and nickel-based battery chemistries with some redesign. Nonetheless, there
are applications that rely on only one type of battery chemistry, e.g., golf cars (which use leadacid), cordless telephones (nickel-metal hydride), and cellular telephones (lithium ion).
In the case of golf cars, the lead-acid chemistry is used because it is the only chemistry
that satisfies the requirements of the application, such as battery voltage and energy
requirements. In that case, those primary capacity and utility characteristics can serve as a proxy
for battery chemistry, and as long as DOE accounts for those other characteristics, it will also
take into account any further effects of battery chemistry. In the case of cordless and cellular
telephones, where the two applications have similar battery voltage and energy requirements,
among others, it appears that the difference in battery chemistry is primarily a function of cost
and historical preference. Cellular telephones tend to be costlier, so manufacturers can better
afford the cost of a more expensive lithium-ion battery. This cost can be accounted for in the
engineering analysis, which examines the cost differences associated with energy efficiency.
Additional differences between nickel- and lithium-based chemistries can be treated
similarly to the differences between slow and fast chargers. Because lithium-based batteries tend
to be more energy-dense and have lower self-discharge, they provide additional consumer utility,
similar to fast chargers. However, like fast chargers, lithium-based BCs also tend to be more
efficient than nickel-based BCs. As a result, even if stringent DOE standards were to limit the
number of nickel-based BCs in the market, doing so would not reduce consumer utility. For the
reasons described above, differences in battery chemistry do not warrant the establishment of

2-23

separate standards under 42 U.S.C. 6295(q), and have therefore not been considered in the
establishment of BC product classes detailed in chapter 3.
Inductive Coupling. Inductively coupled BCs do not use a galvanic (i.e., currentcarrying) connection to transfer power from the DC or AC-line input to the battery; rather, a coil
in the charging cradle transmits power via a magnetic field to a similar coil embedded in the
enclosure containing the remainder of the charger, the integral battery, and the end-use
application. These types of chargers have been typically used for electric toothbrushes, but have
also recently been introduced for use with notebook computers.
AHAM, Philips, and Wahl all commented that DOE should establish a separate product
class for inductively-coupled chargers. (AHAM, No. 16 at p. 4; AHAM, Pub. Mtg. Tr., No. 14 at
p. 124; Philips, No. 22 at p. 2; Wahl, No. 23 at p. 2) They stated consumer benefits of inductive
chargers include safety and cleanliness. By separating the cradle from the rest of the system with
two layers of plastic, and transferring power to the battery through magnetic induction, designers
ensure isolation from wall current. This can be important in a wet environment where most
inductive chargers are used; however, it is worth noting that most BCs provide galvanic isolation
through a transformer enclosed inside a wall adapter. This design decreases the likelihood of
shock by distancing non-isolated line-voltage AC from the user and the wet environment. This
design can be found in BCs for a number of other applications used in wet environments
including shavers, trimmers, and even dental irrigators, which are subject to the same usage
conditions as toothbrushes.
Nonetheless, the commenters imply that metal contacts tend to corrode in a wet
environment (which would potentially cause loss of contact between parts of the system and
require frequent cleaning). Because these events may impact consumer utility, DOE considered
inductive charging for applications used in wet environments when developing product classes in
chapter 3. DOE did not consider separate product classes for inductive chargers used with other
applications (such as notebook computers), since they confer no safety or cleanliness advantages.
Cord-Cordless Operation. Cord-cordless products (e.g., shavers or laptop computers) are
dual-function BC products that can use the wall adapter to both recharge the battery as well as
power the product directly in case the battery is discharged. For example, a user typically
recharges the battery of a shaver through a wall adapter, disconnects the wall adapter, and then
powers the shaver from the battery. However, if the shaver is cord-cordless, the user need not
wait for the battery to recharge and can alternatively power the shaver directly from the wall
adapter.
Because the wall adapter must be able to supply power to the application directly, the
power conversion circuitry has a higher capacity (i.e., output power) than that required to
recharge a similar-sized battery in the absence of cord-cordless operation. Nonetheless, because
this capacity is unrelated to battery charging, DOE views cord-cordless operation as a
characteristic that impacts utility. Both AHAM and Wahl commented that DOE should consider
cord-cordless operation when setting product classes. (AHAM, No. 16 at p. 4; Wahl, No. 23 at
p. 12; Wahl, Pub. Mtg. Tr., No. 14 at p. 291)

2-24

The additional power converter capacity necessary to support cord-cordless operation


will result in higher fixed losses (e.g., higher hysteresis losses for larger transformers).
Specifically, the efficiency of a dual-function system operating far below its design capacity in
battery charging mode will never be as high as that of an optimally scaled single-function
system. This can be seen in the typical relationship between output power and efficiency for an
EPS. When an EPS outputs power above 25 percent of maximum rated output power its
efficiency tends to be much higher than when it outputs power below 25 percent of maximum
rated output power. The reason for this is that at low load (less than 25 percent of maximum
rated output power) an EPSs fixed losses from overhead circuitry are a significant portion of
total losses.
Nonetheless, manufacturers can employ a variety of techniques to improve the efficiency
of a BC designed for cord-cordless operation. This is evident from other BC systems that tend to
operate far below their maximum rated output power. For example, notebook-computer BCs tend
to outperform BCs for other applications such as power tools, despite the fact that notebookcomputer BCs tend to operate at a far lower level of their maximum capacity than do power tool
BCs. Therefore, because there is no systematic efficiency penalty to using a cord-cordless
system, there is no need to set different standards to protect the utility offered by these BCs.
Hence, DOE did not consider cord-cordless operation in the development of BC product classes
presented in chapter 3.
Additional Functionality. Besides cord-cordless operation, manufacturers may include
in the BC product additional features unrelated to battery charging. These features result in
higher energy consumption when compared to similar BCs, leading Wahl to comment that DOE
should consider establishing separate product classes for BCs with additional functionality such
as an embedded radio, etc. (Wahl, No. 23 at p. 2) PG&E commented that standards should
encourage manufacturers to include an off switch for additional functionality, rather than a free
pass to consume excess energy. (PG&E, Pub. Mtg. Tr., No. 14 at p. 294) Similarly, PG&E
suggested that DOE should not allow battery chargers to consume energy for additional
functions. (PG&E, No. 20 at p. 17)
It is possible to categorize BCs with additional functionality into two non-exclusive
groups based on whether the additional functionality is due to connections with other systems
(e.g., cordless telephone base) or due to user interaction (e.g., power tool charger radio).
The first type of additional functionality is exemplified by cordless telephone bases that
monitor the state of the telephone line and/or store voicemail messages. These types of devices
provide an added utility through connection with other systems, e.g., the telephone line. Because
the additional functionality relies on the connection to other parts of the system, manufacturers
can use disconnection (required by the proposed BC active mode test procedure) as a signal to
turn off the additional functionality and reduce power consumption to the level of BCs without
the additional functionality.
The second type of additional functionality is exemplified by a power tool charger radio
that provides an interface for operation by the user. Because this type of device already relies on
users to operate it, a manufacturer should be able to add or repurpose one of the interface
elements to allow a user (and tester) to turn off the additional functionality of the device,

2-25

reducing its power consumption to a level comparable with BCs and EPSs without the additional
functionality.
In either case, the energy consumption of the additional functionality can be substantially
reduced through design features that disable the additional functionality when not needed,
reducing the energy consumption of the BC to the level of similar BCs without additional
functionality. DOE is addressing this issue through its BC active mode test procedure
rulemaking, where it is currently proposing to require that all BC efficiency tests to be conducted
with auxiliary cables disconnected and user switches turned off. Because these test conditions
can be easily detected by the microcontroller operating the BC, designers should be able to
power down the additional functionality. As a result, such BCs likely do not have a higher level
of energy consumption and DOE did not consider additional functionality as a factor in setting
product classes.
DC Input Voltage Type. While not explicitly mentioned in the comments of interested
parties, another potential characteristic that can be used for setting product classes is the input
voltage to the BC. Whereas many BCs receive power from AC mains, DOE has identified
numerous BCs for applications such as portable media players, personal data assistants (PDAs),
and in-car satellite navigation units that draw power from a low-voltage DC source rather than
the high-voltage AC line. In particular, manufacturers of these products do not typically package
a wall adapter, instead expecting users to recharge them using a personal computers USB port or
a cars cigarette lighter receptacle.
Because a rechargeable battery requires direct current (DC) to charge, BCs with linevoltage AC input must include AC-DC conversion; in contrast, BCs with low-voltage DC input
can forgo this step, potentially attaining higher efficiencies. Because input voltage can have a
differential impact on BC efficiency and, furthermore, because input voltage determines where
the BC can be used, which would impact the products utility, DOE is considering using this
characteristic to set product classes.
AC Output. While most BC applications draw direct current (DC) directly from the
battery or through a DC-DC converter, uninterruptible power supplies (UPSs) differ in that they
use an DC-AC converter to provide an AC output from the battery. This difference in design is
required by the end-use applications that are connected to the UPS-devices, such as desktop
computers, monitors, and network equipment. All of these devices typically require line-voltage
AC output.
The additional circuitry required to perform this AC-DC conversion consumes additional
energy even when no applications are connected. Therefore, DOE is considering placing these
BCs in a separate product class.
2.3.2.4 EPS Product Classes under Consideration
For all the reasons discussed above, DOE is considering the EPS product classes listed in
Table 2.2, Table 2.3, Table 2.4, Table 2.5, and Table 2.6. Under these product classes, an EPSs
product class is first determined by whether it is Class A or non-Class A and then by its type of
power conversion, its output voltage(s), and/or its output power. For Class A, medical, and

2-26

MADB EPSs DOE proposes using the same criteria as ENERGY STAR 2.0 to distinguish low
voltage EPSs from basic voltage EPSs. Within each product class, the efficiency standard level
will vary with nameplate output power but not output cord length. Chapter 3 of this preliminary
TSD provides additional detail on the selection product classes.
Table 2.2 Class A EPS Product Classes
Basic Voltage Output
AC-DC Conversion
A1
AC-AC Conversion
A3

Low Voltage Output*


A2
A4

Low voltage output EPSs have nameplate output voltage less than 6 volts and nameplate output current greater than
or equal to 550 milliamps. All other EPSs are basic voltage output.

Table 2.3 Multiple Voltage EPS Product Classes


#
X1
X2

Nameplate Output Power < 100 watts


Nameplate Output Power 100 watts
Table 2.4 High Power EPS Product Classes

#
H1

Nameplate Output Power > 250 watts


Table 2.5 Medical EPS Product Classes
Basic Voltage Output
AC-DC Conversion
M1
AC-AC Conversion
M3

Low Voltage Output*


M2
M4

Low voltage output EPSs have nameplate output voltage less than 6 volts and nameplate output current greater than
or equal to 550 milliamps. All other EPSs are basic voltage output.

Table 2.6 MADB EPS Product Classes


Basic Voltage Output
AC-DC Conversion
B1
AC-AC Conversion
B3

Low Voltage Output*


B2
B4

Low voltage output EPSs have nameplate output voltage less than six volts and nameplate output current greater
than or equal to 550 milliamps. All other EPSs are basic voltage output.

2.3.2.5 BC Product Classes Under Consideration


Based on the above discussion of potential product class criteria for BCs, DOE took into
account the following capacity- and utility-related characteristics when setting product classes:
output power;

battery voltage;

battery capacity;

battery energy;

inductive charging;

input voltage (line AC or low-voltage DC); and

AC output.

2-27

Following additional analysis, detailed in chapter 3 and summarized below, DOE decided
to base its product classes for the preliminary analysis on battery energy and battery voltage, as
well as input voltage (line AC versus DC) and additional functionality (inductive charging and
AC output).
Although output power is the BC characteristic most indicative of BC capacity, battery
voltage, capacity, and energy all impact output power and all relate to BC utility. Of these,
battery energy has both the biggest impact on output power and is also the most meaningful
indicator of utility, measuring the total useful work that can be performed by a charged battery.
Consequently, DOE selected this characteristic to help establish product classes in the
preliminary analysis.
However, as mentioned in section 2.3.2.1 above, the battery voltage provides an
additional measure of utility by constraining the end-user applications that a particular battery
can power. Because of its importance to consumers and impact on BC efficiency, DOE selected
battery voltage for setting product classes in the preliminary analysis. Additional characteristics
selected for setting product classes included inductive charging, input voltage, and AC output.
Because there are many possible combinations of battery voltage and energy, DOE
decided to group BCs into several product classes based on their underlying topology, or design.
BCs for low-energy batteries typically use line-frequency or flyback topologies; BCs for
medium-energy batteries typically use forward or half-bridge topologies; while BCs for highenergy batteries typically use full-bridge or ferro-resonant topologies. These underlying
topologies impact the technology options that can be used for improving the energy efficiency of
the BC, such that an analysis on a BC of a particular topology can be reliably applied to others of
that same topology to the entire product class.
While battery voltage also impacts the choice of technology options, its impact is not as
great as battery energy. To address this impact, DOE further divided some BCs into high-, low-,
and where applicable, medium-voltage product classes. The product classes appear in Table 2.7.
DOE did not separate certain classes by voltage for which only a single application exists on the
market because those classes each have only one voltage.

2-28

Table 2.7 BC Product Classes Used in the Preliminary Analysis.


Product Class
#
Description
Inductive
Low Energy,
1
Connection
Inductive
Low Energy,
<4 V
2
< 100
Low Voltage
Wh
Low Energy,
410 V
3
AC
Med. Voltage
In,
Low Energy,
>10 V
4
DC
High Voltage
Out
Med. Energy,
<20 V
5
Low Voltage
100
3000 Wh
Med. Energy,
20 V
6
High Voltage
> 3000 Wh

<9 V Input

9 V Input

AC Output from
Battery

10

DC In,
DC Out

AC In,
AC Out

High Energy
Low Energy,
Low Voltage DC
Input
Low Energy,
High Voltage DC
Input
Low Energy,
AC Output

Example Applications
Toothbrushes
Telephones
Cameras and Small
Tools
Laptops and Large
Tools
Marine Chargers,
Wheelchairs
Electric Bikes,
Lawnmowers
Golf Cars
USB Chargers

Car Chargers
Uninterruptible Power
Supplies

2.3.3 Technology Assessment


In the technology assessment, DOE identifies technology options that appear to be a
feasible means of improving product efficiency. This assessment provides the technical
background and structure on which DOE bases its screening and engineering analyses. The
following discussion provides an overview of the salient aspects of the technology assessment,
including issues on which DOE seeks public comment. Chapter 3 of the preliminary TSD
provides detailed descriptions of the basic construction and operation of BCs and EPSs, followed
by a discussion of technology options to improve their efficiency and power consumption in
various modes.
2.3.3.1 EPS Efficiency Metrics
On December 8, 2006, DOE codified a test procedure final rule for EPSs in Appendix Z
to Subpart B of 10 CFR Part 430 (Uniform Test Method for Measuring the Energy
Consumption of External Power Supplies.) 71 FR 71340. DOEs test procedure, based on the
California Energy Commission (CEC) EPS test procedure, yields two measurements: activemode efficiency and no-load-mode (standby-mode) power consumption.

2-29

Active mode conversion efficiency is the ratio of output power to input power. The DOE
test procedure averages the efficiency at four loading conditions25, 50, 75, and 100 percent of
maximum rated output currentto assess the performance of an EPS when powering diverse
loads. The test procedure also measures the power consumption of the EPS when disconnected
from the consumer product, which is termed no-load power consumption. Because both the
average efficiency and no-load power consumption affect the energy consumption of the EPS,
DOE developed CSLs for the engineering analysis that are matched pairs of limits on both
metrics simultaneously.
To develop the analysis and ultimately to set standards for EPSs, DOE can either
combine the test values into a single metric, such as annual energy consumption, or it can
maintain two separate metrics. For this preliminary analysis DOE has chosen to evaluate EPSs
using the two metrics separately. DOE believes this approach is the most appropriate way to
characterize EPSs because these products have a wide variety of usage profiles that would affect
any weighting of average efficiency and no-load power. Adopting an integrated approach would
depart from the precedent established by EISA 2007, possibly allowing for backsliding
because a combined metric would limit the overall energy consumption of the EPS, which could
allow for one metric to be below EISA standards if the other metric sufficiently exceeded
standards. Further, this approach is important because the cost estimates in the engineering
analysis apply to the EPS as a whole. . DOE created matched pairs of efficiency to evaluate
CSLs, test data, and manufacturer data. In the matched pairs approach EPS power consumption
improves either through higher active-mode efficiency, lower no-load mode power consumption,
or both.
On April 2, 2010, DOE proposed a test procedure for multiple-voltage EPSs that would
yield five values: no-load power consumption as well as efficiency at 25, 50, 75, and 100percent
of maximum load. 75 FR 16958, 16974. DOE proposes to integrate these four efficiency values
into an average efficiency metric in the standard. Specifically, DOE proposes to use the no-load
power consumption and an average of the four efficiency values as metrics to set candidate
standard levels for multiple-voltage EPSs, similar to single-voltage EPSs. Alternatively, DOE is
also considering approaches to integrate the four active mode measurements in a way to help
better indicate the efficiency of multiple-voltage EPSs during typical use. For instance, DOE
could average only the efficiency values at 50 percent and 75 percent. See Microsoft,
EERE2008BTTP0004 No.8 at p. 1-2. Overall, DOE faces the same challenge with multiplevoltage EPSs as it does with single-voltage EPSs: products change continuously, which means
the test metric should be robust enough to be valid to all applicable products. As noted in the
executive summary, DOE is seeking stakeholder comment on alternative methods for using the
five metrics to set candidate standard levels for EPSs.
2.3.3.2 EPS Technology Options
In the framework document, DOE requested comments on a list of preliminary
technology options that DOE intended to study further. For EPSs, these technology options
included:
Improved Cores

2-30

Switched-Mode Power Supply


Low-Power Integrated Circuits
Schottky Diodes and Synchronous Rectification
Low-Loss Transistors
Low-Loss Snubbers
Interested parties did not provide comment on the technology options proposed in the
framework document nor suggest additional technology options for DOE to consider. Additional
information on these technology options is available in chapter 3.
2.3.3.3 BC Efficiency Metrics
As described in section 2.1.2.2, DOE has adopted a test procedure for BCs in appendix Z
to subpart B of 10 CFR part 430. In a related rulemaking, DOE is currently updating this test
procedure to accommodate measurement of active mode energy consumption, and the proposed
test procedure will return four metrics, related to each mode of operation of a BC: active,
maintenance, standby, and off.
At the framework document public meeting, DOE expressed its intention of weighting
these mode-specific metrics using a product-class average usage profiled to calculate the annual
energy consumption of a BC. These weights would be calculated during the standard-setting
stage of the rulemaking.
PTI and AHAM commented that a BC standard should incorporate a usage profile. (PTI,
Public Meeting Transcript, No. 14 at p. 249; AHAM, Public Meeting Transcript, No. 14 at
p. 251) PG&E disagreed, noting that while duty cycle estimates are needed to calculate energy
use and savings, they should not be used to create a standard metric. (PG&E, No. 20 at p. 9)
Furthermore, standards expressed in terms of separate metrics would ensure that all operational
modes are efficient. (PG&E, No. 20 at p. 6) Finally, PG&E believed that it would be difficult to
develop appropriate usage profiles as there is insufficient data to arrive at a realistic and
creditable understanding. (PG&E, Pub. Mtg. Tr., No. 14 at pp. 161,and 159; PG&E, No. 20 at
p. 15). PG&E stated that DOE should rather defer the creation of appropriate usage profiles to
future rulemakings once more data have been obtained. (PG&E, Pub. Mtg. Tr., No. 14 at p. 178)

DOE is obligated to create an integrated measurement of overall energy consumption, if


possible. Section 325 of EPCA, as modified by EISA 2007, requires the results of a standby
measurement to be:

The calculation of product-class-average usage profiles and the calculation of typical energy consumption is
described in detail in the energy use and end-use load characterization (chapter 7).

2-31

integrated into the overall energy efficiency, energy consumption, or other energy descriptor
for each covered product unless the Secretary determines that
(i) the current test procedures for a covered product already fully account for and
incorporate the standby mode and off mode energy consumption of the covered product;
or
(ii) such an integrated test procedure is technically infeasible for a particular covered
product, in which case the Secretary shall prescribe a separate standby mode and off
mode energy use test procedure for the covered product, if technically feasible. (42
U.S.C. 6295(gg)(2)(A))
Combining the energy consumption of BCs in each modeincluding standbyinto a
single energy consumption metric through usage profiles, as described above, may allow DOE to
establish such an integrated metric.
Furthermore, DOE acknowledges that, in contrast to other consumer products
rulemakings, where the number of distinct usage profiles is small, there are many BC end-use
products with many different usage profiles. Thus, determining usage profiles representative of
the wide variety of applications, as well as any future applications, could be difficult because of
this wide variation. Nonetheless, usage profiles must be usedas PG&E agrees to calculate
the energy consumption associated with a potential standard in order to evaluate its impacts on
energy savings. Therefore, regardless of their limitations, their incorporation in the standards
analysis is unavoidable.
Finally, AHAM commented that an integrated metric would allow manufacturers to
adjust energy consumption in each mode to accommodate customer usage, citing previous DOE
precedent in the case of clothes washers. (AHAM, No. 16 at p. 4) DOE agrees that such an
integrated metric would give manufacturers some flexibility to improve the efficiency of their
products in a cost-effective fashion by enabling them to focus on decreasing energy
consumption in the modes where they will achieve the most efficiency gains.
However, to best reflect actual usage would require setting separate standards for each
application that uses BCs. Because of the complexity associated with the large number of
applications, DOE has elected to group the usage profiles according to the product classes
described earlier in section 2.3.2. While these aggregate usage profiles may not be an exact
match for every application in the product class, they incorporate some of the features of each,
and are more representative than PG&Es proposal not to use usage profiles at all.
The test procedure that DOE ultimately adopts would likely require energy consumption
measurements in each mode as presented in the agencys test procedure proposal. See 75 FR
16958. These modal metrics would then be multiplied by the time spent in each mode, as
specified by the above-described average usage profiles. These metrics would be contained in the
standards section of the Code of Federal Regulations (CFR), similar to those that apply to
dishwashers. See 10 CFR 430.23(c).

2-32

2.3.3.4 BC Technology Options


Since most consumer BCs contain an AC/DC power conversion stage, similar to that
found in an external power supply, all of the technology options discussed above in section
2.3.3.2 also apply to BCs. The technology options used to decrease EPS no-load power will
impact energy consumption of BCs in no-battery and maintenance modes (and off mode, if
applicable), while those used to increase EPS conversion efficiency will impact energy
consumption in active and maintenance modes.
Additional technology options specific to BCs that DOE presented in the framework
document included:
Termination
Elimination/Limitation of Maintenance Current
Elimination of No-Battery Current
Phase Control to Limit Input Power
Several interested parties commented on these technology options. Micro Power
Electronics noted the difficulty of implementing an AC power disconnect in a multi-bay or
multi-port BC (i.e., one that charges several batteries independently). Such a disconnecta
potential way to eliminate no-battery currentwould be difficult to implement because the
charge controller must remain powered to detect when a battery is inserted into the charge bay.
(MPE, No. 5 at p. 1) While DOE agrees that disconnecting the BC from the AC line may be
more complicated in a multi-port charger, it is still a viable technology option, and is including it
in the technology assessment.
PG&E noted that the ability of the BC to detect a full battery and subsequently reduce
current is a key way to reduce BC energy consumption. (PG&E, No. 13 at p. 10; PG&E, Pub.
Mtg. Tr., No. 14 at p. 53 and 303) Whereas in the previous comment disconnection of the
charger from the AC line is a technology option that decreases no-battery power, this comment
pertains to maintenance mode. When the battery is fully charged, and the BC enters maintenance
mode, BCs that terminate reduce the output current to the battery and thereby the input power to
the charger. Without termination, this excess power is given off as heat in the battery.
Termination is still a viable technology option, and DOE will evaluate it in the technology
assessment.
PG&E suggested that DOE consider a capacitive reactance circuit as a replacement for
line-frequency transformers in slow chargers. (PG&E, No. 20 at p. 6; Pub. PG&E, Mtg. Tr., No.
14 at p. 304) All BCs that connect to mains must convert higher-voltage AC mains power to
lower-voltage DC power, because batteries only accept DC power. Whereas conventional power
converters use a transformer to lower the voltage, a capacitive reactance circuit is another,
potentially more efficient way to convert to a lower voltage.
However, capacitive reactance circuits cannot be used in all applications because they are
typically low-current and take multiple hours to charge a battery. Test results of a highly efficient
2-33

capacitive reactance charger submitted by PG&E reveal maintenance mode power of 0.1 watts
and no-battery mode power of 0 watts. However, the initial charge mode power for a 7.1 watthour battery was 0.3 watts. Even if the battery charging system were 100 percent efficient, it
would take approximately 24 hours to charge the battery, and in reality the test was run for
96 hours (i.e., 4 days) to ensure a full charge.
Furthermore, because it does not use a transformer, the capacitive reactance circuit does
not provide galvanic isolation between input and output, meaning that the output of the circuit
and any other electrical components connected to the output (e.g., battery terminals) must be out
of reach of the user to prevent shock. Use of this circuit would also require adequate fireproofing
of the end-use product if the battery is integral, as in the case of cordless telephones or handheld
vacuum cleaners.
Because the longer charging times and requirements to fireproof the end-use product
enclosure, capacitive reactance circuits may have adverse impacts on product utility, which is
one of the criteria used in the screening analysis. Detailed discussion of this technology option
and its applicability to each product class can therefore be found in chapter 4.
PG&E presented a further list of technology options for decreasing the energy
consumption of battery chargers:
Improve power supply efficiency
Reduce power supply no-load power
Maintenance by periodic topping-off charge
Switched-mode charge controller
Automatic battery-connected on-off switch
Improved electronic/magnetic components beyond the power supply
Lower charging current or higher voltage (PG&E, No. 13 at p. 10)
DOE included the above technology options in the technology assessment (see chapter 3). The
final option suggested by PG&Elowering charging current or increasing voltagewas
screened out on the grounds that it would negatively impact consumer utility as it would require
product redesign, with potentially detrimental results (see chapter 4).
2.4

SCREENING ANALYSIS

This section of the preliminary TSD focuses on comments from interested parties
pertaining to the screening analysis as well as technology options that were screened out due to
their failure to comply with one or more of the four screening criteria. Those technology options
that were not screened out are discussed in chapter 4. After DOE identified the technologies in
the technology assessment that could potentially improve the energy efficiency of BCs and EPSs,
2-34

DOE conducted the screening analysis. The purpose of the screening analysis is to evaluate the
technology options that improve the efficiency of equipment to determine which options to
consider further and which options to screen out. DOE consults with industry, technical experts,
and other interested parties in developing a list of design options. DOE then applies the
following set of screening criteria to determine which design options are unsuitable for further
consideration in the rulemaking (10 CFR Part 430, subpart C, appendix A at 4(a)(4) and 5(b)):
Technological Feasibility - DOE will consider technologies incorporated in commercial
equipment or in working prototypes to be technologically feasible.
Practicability to Manufacture, Install, and Service - If mass production of a technology
in commercial equipment and reliable installation and servicing of the technology could
be achieved on the scale necessary to serve the relevant market at the time of the
compliance date of the standard, then DOE will consider that technology practicable to
manufacture, install, and service.
Adverse Impacts on Equipment Utility or Equipment Availability - If DOE determines a
technology to have a significant adverse impact on the utility of the equipment to
significant subgroups of consumers, or to result in the unavailability of any covered
equipment type with performance characteristics (including reliability), features, sizes,
capacities, and volumes that are substantially the same as equipment generally available
in the United States at the time, it will not further consider this technology.
Adverse Impacts on Health or Safety - If DOE determines that a technology will have
significant adverse impacts on health or safety, it will not further consider this
technology.
For a complete discussion of the screening analysis, refer to chapter 4.
2.4.1 Technology Options Screened Out
PG&E commented that switched mode power supplies can power all non-medical
consumer applications and should not be screened out. (PG&E, No. 20 at p. 8) In its comment,
PG&E was alluding to the concern that switched-mode power supplies can have more electrical
noise than linear-regulated power supplies. This noise can be transmitted to nearby circuits either
through direct connections (through data lines, for example) or transmitted through the air (with
power cables acting as antennas). However, technology advances for switched-mode power
supplies have greatly reduced electrical noise issues, which would enable switched-mode power
supplies to replace linear-regulated power supplies. At this time, DOE is unaware of cases
where linear-regulated power supplies are the only technology option. In light of these advances,
DOE did not screen out switched-mode power supplies or any other EPS technology option.
AHAM stated that an energy conservation standard should not impact the safety of a
product, particularly those products used in wet environments. (AHAM, Pub. Mtg. Tr., No. 14 at
p. 43) DOE agrees that energy conservation standards should not impact the safety of a consumer
product, which is one of the criteria DOE uses when evaluating technology options in its
screening analysis. However, as mentioned earlier, any power conversion technology that
providing galvanic isolation should be able to meet the safety requirements of BCs used in wet
environments. Additionally, cleanliness, another concern is addressed in the screening analysis in

2-35

chapter 4 through the application of the adverse impacts on product utility criterion. See supra,
section 2.3.2.3.
Similarly, the safety criterion was also used to screen out the capacitive reactance
technology option, which involves the use of a capacitor to provide voltage transformation
without galvanic isolation.
In summary, DOE screened out the following BC technology options:
Non-inductive chargers for use in wet environments
Capacitive reactance
Lowering charging current or increasing battery voltage
2.5

ENGINEERING ANALYSIS

In the engineering analysis (detailed in chapter 5), DOE develops a relationship between
the manufacturer selling price (MSP) and increases in BC and EPS efficiency. The efficiency
values range from that of a typical BC or EPS sold today (i.e., the baseline) to the maximum
technologically feasible efficiency level. At each efficiency level examined, DOE determines the
consequent MSP, a relationship referred to as a cost-efficiency curve.
Below, DOE presents cost-efficiency curves for BCs and Class A EPSs. For non-Class A
EPSs, DOE explains how it will develop cost efficiency curves based on the Class A EPS
analysis and the non-Class A determination analysis.
2.5.1 Approaches to Conducting the Engineering Analysis
DOE structured its engineering analysis around two methodologies: (1) test and
teardowns, which involves testing products for efficiency and determining cost from a detailed
bill of materials derived from tear-downs and (2) the efficiency-level approach, whereby
manufacturers provide and explain their costs of achieving increases in energy efficiency at
discrete levels of efficiency.
Interested parties expressed concern with the validity of the BC data obtained from
manufacturer interviews in the efficiency-level approach. PG&E. commented that manufacturer
interviews may provide incorrect costs as many manufacturers of battery-powered products
subcontract the manufacturing of the BC and may not be familiar with the design details and
costs. (PG&E, No. 20 at p. 12) PG&E. are correct in noting that BCs are typically purchased by
the end-use product manufacturer or original equipment manufacturer (OEM) from an external
supplier as a completed subassembly. Nonetheless, these manufacturers exercise strict control
over the specification and design of the BCs, and are knowledgeable regarding their cost and
achievable efficiencies. Furthermore, DOE supplemented the efficiency information obtained
from manufacturers with teardowns conducted on BCs currently in the market.

2-36

PG&E noted that there are inexpensive products with very efficient BCs and expressed
its belief that BC technology must, in turn, be inexpensive. (PG&E, Pub. Mtg. Tr., No. 14 at
p. 58) DOE acknowledges PG&Es comment that some applications with BCs are inexpensive
and that their components are likely to be similarly inexpensive. Nonetheless, DOE is using the
testing and teardown method to calculate costs for all BCs, which will more directly demonstrate
the BCs costs. When analyzing the cost of each CSLwhether based on existing or theoretical
designsDOE differentiates the cost of the BC or EPS from the cost of the associated end-use
product.
2.5.2 Engineering Analysis for Class A External Power Supplies
2.5.2.1 Class A EPS Representative Product Classes and Representative Units
As discussed earlier in section 2.3.2.4, DOE divided Class A EPSs into four product
classes for the preliminary analysis, following an examination of EPS technologies in the market
and approaches used in other energy-efficiency programs. There were no comments from
interested parties regarding Class A EPS product classes.
Further examination of EPS units in the market led to the selection of one of the four
product classes for further analysis as the representative product class. This class collectively
constitutes the majority of EPS shipments and national energy consumption. DOE analyzed this
class through testing, teardowns, and interviews with manufacturers. By focusing the analysis on
the representative product class, DOE can provide the most accurate standards analysis for the
applications that have the largest impact on users (due to high shipments, high per-unit energy
consumption, or both) or that have the largest impact on the nation (due to high shipmentweighted energy consumption). Table 2.8 presents representative product class A1 in the context
of the EPS product classes presented in Table 2.2.
Table 2.8 Class A EPS Representative Product Class and Scaled Product Classes
Basic Voltage Output Low Voltage Output *
AC-DC Conversion
AC-AC Conversion

A1 (representative)
A3 (scaled)

A2 (scaled)
A4 (scaled)

* Low voltage output EPSs have nameplate output voltage less than six volts and nameplate output current greater
than or equal to 550 milliamps. All other EPSs are basic voltage output.

DOE subsequently focused its analysis on four representative units within the
representative product class (A1), presented in Table 2.9. Because results from the analysis of
these representative units would later be extended to additional EPSs, they were selected from
high-volume and/or high-energy-consumption applications that use EPSs that are typical across
EPSs in product class A1. Interested parties did not comment on the selection of representative
units for analysis. Chapter 5 presents the test results used by DOE in its analysis.

2-37

Table 2.9 Class A EPS Representative Units


Representative Nameplate Output Power
Unit
[watts]
1
2.5
2
18
3
60
4
120

Nameplate Output Voltage


[volts]
5
12
15
19

Example
Application
Mobile phone
Modem
Laptop Computer
Laptop Computer

2.5.2.2 Class A EPS Candidate Standard Levels Analyzed for Product Class A1
In its engineering analysis for Class A EPSs, DOE examined the impacts on cost of
improving the efficiency of each of the representative units to evaluate the impact of potential
energy efficiency standards. DOE performed this analysis by evaluating test results and
manufacturer data, indicating possible steps to increase the efficiency of currently-available
EPSs. To reflect these general steps in improving efficiency, DOE created several candidate
standard levels (CSLs), as sets of equations, which represent increasingly stringent standards
with higher-numbered CSLs. These CSLs are summarized for each representative unit in Table
2.10, expressed in terms of average efficiency in active mode and power consumption in no-load
mode.
Table 2.10 Summary of Class A EPS Candidate Standard Levels of Efficiency
Number Reference
Basis
CSL 0
EISA 2007
EISA 2007 equations for efficiency and no-load power
ENERGY STAR
ENERGY STAR 2.0 equations for efficiency and no-load
CSL 1
2.0
power
CSL 2
Intermediate
Curve fit to data points
CSL 3
Best in Market
Curve fit to data points
CSL 4
Max Tech
Curve fit to data points
CSLs are generally based on (1) other voluntary specifications or mandatory standards
that cause manufacturers to develop products at particular efficiency levels; (2) the most efficient
products available in the market; and (3) the maximum technologically feasible (max tech)
level.e
DOE considered the current DOE standard level as the baseline CSL. EPCA, as amended
by EISA 2007, determines EPS minimum efficiency and maximum no-load power consumption
as a function of nameplate output power, shown in Table 2.11 and Table 2.12 , respectively. For
this analysis, the baseline efficiency level, CSL 0, for each representative unit is the EISA level
at the corresponding nameplate output power.

The max-tech level represents the most efficient design that is commercialized or has been demonstrated in a
prototype with materials or technologies available today. Max tech is not constrained by economic justification
apart from the requirement that there be more than one unique way to achieve it, per the discussion in 2.5.2.2. It is
typically the most expensive design option considered in the engineering analysis.

2-38

Table 2.11 Baseline (CSL 0) EISA 2007 Active-Mode Efficiency for Product Class A1
Nameplate Output
Active Mode Required Efficiency
(decimal equivalent of a percentage)
<1 Watt
0.5 times the nameplate output
1 to not more than 51 Watts The sum of 0.09 times the natural logarithm
of the nameplate output and 0.5
>51 Watts

0.85

Table 2.12 EISA 2007 Standard Equation for Maximum No-Load Mode Power
Consumption
Nameplate Output

Maximum Consumption

Not more than 250 Watts 0.500 watts


ASAP and PG&E asked DOE to include the ENERGY STAR EPS Version 2.0 as a CSL.
(ASAP, Pub. Mtg. Tr., No. 14 at p. 40; PG&E, Pub. Mtg. Tr., No. 14 at p. 63) To that end when
creating CSL 1, DOE considered the ENERGY STAR 2.0 specification. Similar to EISA, the
ENERGY STAR 2.0 level is a function of nameplate output power, shown in
Table 2.13 and Table 2.14. In addition to output power, ENERGY STAR 2.0 also considers the
type of output power (DC or AC) and the output voltage. DOE chose CSL 1 to be the same as
the ENERGY STAR 2.0 level for AC-DC basic output voltage EPSsf. DOE chose the divisions
in the CSL 1 equations to be consistent with the divisions between the ENERGY STAR 2.0
equations. The two sets of equations do not correspond exactly, however. Specifically, the
divisions between the EISA equations are at 1 watt and 51 watts (shown in Table 2.11), but the
divisions between the ENERGY STAR equations are at 1 watt, 49 watts, and 50 watts (shown in
Table 2.13 and Table 2.14).
Table 2.13 ENERGY STAR 2.0 Equation for Minimum Active-Mode Efficiency for Basic
Output Voltage EPSs
Nameplate Output Power (Pno) Minimum Average Efficiency in Active Mode
(expressed as a decimal)
0 to 1 watt
0.480 * Pno + 0.140
> 1 to 49 watts

[0.0626 * ln (Pno)] + 0.622

> 49 watts

0.870

ENERGY STAR documents use the term standard voltage. Because the word standard is a term of art within
DOE documents, DOE instead uses the term basic voltage to reduce possible confusion.

2-39

Table 2.14 ENERGY STAR 2.0 Standard Equation for No-Load Mode Power
Consumption
Nameplate Output Power (Pno) Maximum Power in No-Load
Ac-Dc EPS
0 to 50 watt
0.300 watts
0.500 watts

> 50 watts

DOE created CSL 2 to be an intermediate level between the Energy Star (CSL 1) and best
in market (CSL 3) levels. The specific combination of no-load power consumption and average
efficiency for CSL 2 was chosen to optimize the tradeoff between cost and efficiency between
CSL 1 and CSL 3. DOE evaluated each representative unit individually and developed sets of
efficiency and no-load power pairings between CSL 1 and CSL 3. Using these data, DOE
estimated their resultant unit energy consumptions (UECs) (see chapter 7 for details on
calculating UECs) and compared the UECs against their associated MSPs, which were calculated
from the aggregation of manufacturer MSP data (see chapter 5 for details on MSP aggregation).
Subsequently, the final efficiency and no-load pairing used to characterize CSL 2 was selected
by examining which of the prospective pairings had the highest weighted-average savings in the
LCC analysis.
DOE then used the CSL 2 pairings for the four representative units to create equations for
average efficiency and no-load power by curve-fitting the efficiency characteristics for CSL 2
(see chapter 5 for details on curve fits). For both the average efficiency and no-load power CSL
equations, DOE used equations similar to those for CSL 1, involving linear and logarithmic
terms in the nameplate output power. DOE chose the divisions at 1 watt, 49 watts, and 50 watts
in the CSL 2 equations to ensure consistency with the nameplate output power divisions between
the equations for CSL 1.
Table 2.15 CSL 2 Intermediate Active-Mode Efficiency for Product Class A1
Nameplate Output Power Minimum Average Efficiency in Active Mode
(expressed as a decimal)
0.49
*
Pout
+ 0.15
0 to 1 watt
0.0701 * ln(Pout) - 0.0011 * Pout +0.647
> 1 to 49 watts
> 49 watts

0.870

Table 2.16 CSL 2 Intermediate No-Load Mode Power Consumption for Product Class A1
Nameplate Output Power Maximum Power in No-Load
0.200 watts
0 to 50watt
0.230 watts
> 50 watts
CSL 3 reflects the most efficient product in the market (best in market or BIM), which
DOE identified using test results of popular products identified in a market survey, models

2-40

identified in the ENERGY STAR 2.0 databaseg, and units available through EPS distributor
websites. CSL 3 corresponds to the combination of average efficiency and no-load power
consumption of the most efficient EPS units that DOE tested. For those units, DOE created the
equations for average efficiency and no-load power using curve fits of the test results data among
the four representative units (see chapter 5 for details on curve fits). For both the average
efficiency and no-load power CSL equations, DOE used equations similar to those for CSL 1,
involving linear and logarithmic terms in the nameplate output power. DOE chose the divisions
at 1 watt, 49 watts, and 50 watts in the CSL 3 equations so that they were consistent with the
nameplate output power divisions between the equations for CSL 1.
Table 2.17 CSL 3 Best in Market Active-Mode Efficiency for Product Class A1
Nameplate Output Power Minimum Average Efficiency in Active Mode
(expressed as a decimal)
0.5
*
Pout
+
0.16
0 to 1 watt
0.071 * ln(Pout) - 0.0014 * Pout +0.67
> 1 to 49 watts
> 49 watts

0.880

Table 2.18 CSL 3 Best in Market No-Load Mode Power Consumption for Product Class A1
Nameplate Output Power Maximum Power in No-Load
0.100
0 to 50watt
0.210
> 50 watts
Finally, DOE determined CSL 4, the maximum technologically feasible (max-tech)
efficiency level, as required by section 325(o) of EPCA. (42 U.S.C. 6295(o)). Earthjustice
commented that DOE cannot legally ignore the evaluation of BC or EPS technologies considered
to be proprietary or prototypes when developing the max tech CSL. (Earthjustice, No. 18, at p. 1)
The agency notes that in determining whether standards would meet the statutorily-mandated
criteria of being economically feasible, the agency must examine the competitive impacts of its
choice of standards. By selecting a path that would force regulated entities to adopt a single,
proprietary technology as the only feasible means of satisfying the standard, DOE would run
afoul of its obligations under 42 U.S.C. 6295(o) to consider the potential for lessened
competition. The selection of such a path would not be likely to withstand the required scrutiny
from the U.S. Department of Justice provided under the statute. (See 42 U.S.C.
6295(o)(2)(B)(i)(I) through (VII))
To determine the max tech CSL, DOE interviewed manufacturers of EPSs, manufacturers
of integrated circuits for EPSs, and manufacturer of applications that use EPSs. All of the
manufacturers interviewed indicated how EPSs could achieve efficiencies higher than currently
exist in the market and at what cost, Chapter 5 details this information. The accuracy of this
information was independently verified by subject matter experts (SMEs) consulted by DOE.
The SMEs believed that the data was reasonable, except for the max-tech value for the 2.5W
g

Taken in September 2009 from the ENERGY STAR 2.0 database at


http://www.energystar.gov/index.cfm?fuseaction=find_a_product.showProductGroup&pgw_code=ACD

2-41

EPSs. The SMEs believe that those EPSs may be able to achieve a max tech efficiency of 80%
rather than the 74.0% efficiency derived from manufacturers. The aggregated responses from
manufacturers to that assertion are discussed in chapter 5. Because the 80% efficiency figure
from SMEs had no corresponding cost data, DOE used a max-tech efficiency value of 74.0% to
characterize CSL 4, because DOE had corresponding cost data at that efficiency level.
DOE created the CSL 4 equations for average efficiency and no-load power using curvefits of the aggregated manufacturer data (see chapter 5 for details on curve fits). DOE created the
equation for no-load power based on a curve fit of the no-load power among the four
representative units. For both the average efficiency and no-load power CSL equations, DOE
used equations similar to those for CSL 1, involving linear and logarithmic terms in the
nameplate output power. DOE chose the divisions at 1 watt, 49 watts, and 50 watts in the CSL 4
equations so that they were consistent with the nameplate output power divisions between the
equations for CSL 1.
Table 2.19 CSL 4 Max Tech Active-Mode Efficiency for Product Class A1
Nameplate Output Power Minimum Average Efficiency in Active Mode
(expressed as a decimal)
0.52* Pout + 0.17
0 to 1 watt
> 1 to 49 watts
> 49 watts

0.079 * ln(Pout) - 0.0014 * Pout +0.67


0.910

Table 2.20 CSL 4 Max Tech No-Load Mode Power Consumption for Product Class A1
Nameplate Output Power Maximum Power in No-Load
0.062
0 to 50watt
0.165
> 50 watts

2.5.2.3 Class A EPS Teardown Analysis


For each representative unit, DOE analyzed the cost and efficiency of three EPS units
that it purchased to characterize the baseline efficiency (CSL 0), Energy Star (CSL 1), and best
in market (CSL 3) (except for the 120-watt representative unit; for that, DOE purchased two test
units). To ensure that the results were comparable between units, DOE defined specific criteria
for the representative unit for nameplate output power, nameplate output voltage, and output
cord length. All of the EPS test units were chosen so that their specifications were close to or the
same as the representative unit specifications. For the EPSs that differed slightly, DOE scaled
their test results as specified in chapter 5. DOE then obtained an estimate of the costs for each of
these EPS test units.
Because detailed EPS component costs at large volumes are unavailable on the
commercial market, DOE contracted iSuppli Corp., an industry expert in costs of consumer
electronics, to tear down and estimate the materials cost for select units. iSuppli provided DOE

2-42

with the cost for all parts listed in the bill of materials (BOM cost). DOE used that information,
together with the labor costs of assembly, the factory overhead costs, and markup costs to
determine the manufacturers selling price (MSP), as shown in Figure 2.1. These cost estimates
also account for the typical number of units produced by the manufacturer as well as the
manufacturers location.

Full Cost of Product [$]

Production Cost
Direct
Labor

Direct
Materials

Overhead

Non-Production Cost
Selling

General &
Administrative

R&D

Interest

Profit

Figure 2.1 Full Cost of Product: Breakdown of Production and Non-Production Costs
2.5.2.4 Class A EPS Manufacturer Interviews
In 2009 and 2010, on behalf of DOE, Navigant Consulting, Inc. (Navigant Consulting)
interviewed manufacturers of EPSs, integrated circuits for EPSs, and applications that use EPSs
to obtain data on EPS efficiencies and the costs associated with achieving those efficiencies. At
the request of some manufacturers, Navigant Consulting entered into non-disclosure agreements
whereby it can present to DOE general information about the EPS market and technology, but no
confidential data specific to any individual manufacturer. These interviews enabled Navigant
Consulting to obtain general information about the EPS market and technology to conduct the
analysis without attributing any particular data to an individual manufacturer. DOE continues to
seek input from interested parties regarding all aspects of the rulemaking, in particular cost and
efficiency data.
2.5.2.5 Class A EPS Scaling Relationships
DOE developed scaling relationships that it used both to analyze representative unit data
and to scale CSLs from product class A1 to other product classes. In general, the scaling methods
for representative unit data were detailed whereas the scaling methods for CSLs were simpler.
For the representative unit data DOE adopted the more detailed approach because the detailed
methods provide enhanced accuracy, which was paramount, as it was the basis for later analyses,
including the LCC and NIA. For the CSL equations, DOE adopted the simpler approach so that
stakeholders could more easily evaluate the CSLs. As noted in the executive summary, DOE is
seeking stakeholder comment on these approaches and may develop more detailed CSL
equations in response to stakeholder comment.
For the purposes of the preliminary analysis, DOE developed CSL equations that have
the general form Y=aln(Pout) + bPout+c, for each of the nameplate output power segments,
where Y indicates the efficiency or no load-power requirement; Pout indicates the nameplate

2-43

output power; and a, b, and c indicate the specific parameters defined in the respective CSLs. As
explained in chapter 5, DOE has determined that output cord length and output voltage also have
an effect on efficiency. Thus, DOE could have included terms in the CSL equations for output
cord length and output voltage but doing so would have resulted in much more complicated
equations that would have minimally enhanced accuracy. DOE accounts for differences in output
voltage through different product classes, but does not account for differences in output cord
length. For the NOPR, DOE intends to maintain the equations in this form, which is based on
output power exclusively, unless interested parties prompt DOE to change this approach.
Wherever possible, DOE selected test units that had the same characteristics as the
representative units. Similarly, it sought data from manufacturers at the representative unit
values. However, in some cases, test units and manufacturers data were slightly different from
the representative unit specifications. In addition, the teardown and manufacturer cost data had to
be normalized such that the costs included the same assumptions. For those cases DOE
developed detailed scaling relationships so that the data would be most applicable to the
representative unit analysis.
Specifically, the Class A EPS engineering analysis characterizes the cost-efficiency
relationship using average efficiency in active mode, power consumption in no-load mode and
MSP. DOE did not develop scaling relationships for no-load mode power consumption because
it is minimally affected by EPS representative unit characteristics. DOE considered developing a
scaling relationship for cost by production volume. Specifically, for the representative units,
DOE and iSuppli developed estimate production volumes that were on the order of
approximately 1,000,000 units shipped per year. iSuppli indicated that prices did not change
appreciably over that range; thus. scaling by production volume is not necessary.
For the efficiency and MSP characteristics, DOE developed the following scaling
relationships, detailed in chapter 5:
Efficiency by nameplate output power. DOE scaled efficiency by output power
using the EISA 2007 standard and ENERGY STAR 2.0 specification, which
follow the market trend of increasing efficiency with increasing output power.
Efficiency by nameplate output voltage. For EPSs of a given nameplate output
power, lower nameplate output voltage results in higher current-associated losses
and lower efficiency. DOE analyzed units with the same nameplate output power
and different nameplate output voltages to characterize this relationship. DOE
also evaluated a voltage scaling approach based on ENERGY STAR 2.0, for lowvoltage units.
Efficiency by output cord length. Longer output cords provide consumers with the
added utility of being able to operate the product farther from mains. Longer
output cords also have higher losses and are more expensive. DOE analyzed
output cords to characterize this relationship.
Cost by nameplate output power. EPSs with higher nameplate output powers
provide consumers with the added utility of being able to provide power to more
power-demanding products. EPSs with higher nameplate output power also have
more and larger components and, consequently, cost more.

2-44

Cost by output cord length. Longer output cords are more expensive. DOE
analyzed output cords to characterize this scaling relationship.
Markups. Test and manufacturer data costs at different points in the value chain
represent different stages in the sale of an EPS. DOE characterized these markups
in order to compare costs at the same stage in the value chain, namely MSP.
Though in many cases the data did not require applying all of the scaling steps, all the
scaling steps are nonetheless illustrated in the scaling section of chapter 5, in the order that they
would be applied.
2.5.2.6 Class A EPS Relationships Between Cost and Efficiency
Because DOE used two methods to conduct the engineering analysis (the test and
teardown data and aggregated manufacturer data methods), it generated two distinct outputs
one from each methodfor each representative unit.
For the test and teardowns analysis, DOE combined the scaled efficiency test results with
the MSP for the three EPS test units used to characterize the baseline efficiency level, CSL 1,
and CSL 3. The test and teardown analysis method does not include values for CSL 4 because,
by definition, maximum technologically feasible units are typically not available for purchase.
For the manufacturer interview analysis, DOE presented manufacturers with CSL 0,
CSL 1, and CSL 3 as developed in the testing and teardown analysis. DOE asked manufacturers
to develop their own estimate for CSL 4, max tech. For each of these CSLs, the manufacturers
detailed the changes they would make to the EPS design and the associated costs. To maintain
confidentiality, the manufacturers data was aggregated.
In summary, for each representative unit, the manufacturer data for cost versus efficiency
and cost versus no-load power data showed a trend of increasing MSPs with higher CSLs,
meaning that it costs more to manufacture more efficient EPSs. On the other hand, the testing
and teardown curves did not show a clear cost-efficiency relationship. In fact, in many cases, the
CSL 3, best in market unit was the cheapest, and the CSL 0 unit was the most expensive. For all
four representative units, the cost decreased from CSL 1 to CSL 3.
DOE believes that, although it rigorously normalized the test unit data, there may be
factors that affect cost which were not normalized. For example, as detailed in chapter 5,
manufacturers indicate that reliability is another factor that greatly impacts EPS cost and that it is
difficult to determine an EPSs reliability without access to its specification sheet. Consequently,
DOE believes the testing and teardown results to be inconclusive. Therefore, DOE plans to
evaluate only the manufacturer data in its UEC, LCC, and NIA analyses for Class A EPSs. For
the three representative units whose testing- and teardown-based cost-efficiency curves have
CSL 3 as the least expensive CSL, the LCC for CSL 3 would be positive. For the NOPR, DOE
intends to conduct another round of testing and teardowns for Class A EPSs that attempts to
control for additional factors. In particular, DOE intends to attempt to control the data for
reliability by selecting EPSs with similar end-use applications of similar quality.

2-45

2.5.2.7 CSLs for Class A EPS Product Classes Not Directly Analyzed (A2, A3, and
A4)
DOE identified and selected basic voltage AC/DC EPSs as the representative product
class on which to concentrate its analytical effort. DOE chose this representative product class
primarily due to its high market volumes and its ability to characterize the EPS market. DOE
followed a two-step process to scale the CSLs from the representative units to product classes.
First, DOE scaled from the four representative units with specific nameplate output power values
to create a function that addresses all nameplate output power values in the AC-DC basic voltage
EPS product class. DOE based this extrapolation on existing relationships defined by the EISA
2007 standard for EPSs and the ENERGY STAR 2.0 specification for EPSs. Both the standard
and specification use an equation to establish a correlation between an EPSs nameplate output
power and its efficiency.
Next, DOE used the representative unit data and the CSL equations from product class
A1 to create equations for the other three Class A EPS product classes. Because of similarities
between Class A EPSs, DOE only needed to create two sets of additional equations to apply to
the three non-analyzed product classes: a set of low-voltage efficiency equations and a set of
AC/AC no-load power equations. Low-voltage EPSs (product classes A2 and A4) tend to have
lower efficiencies than basic voltage EPSs (product classes A1 and A3), as explained in detail in
chapter 5. In addition, DOE believes that AC/AC EPSs (product classes A3 and A4) tend to have
higher no-load power consumption than AC/DC EPSs (product classes A1 and A2) because the
ENERGY STAR 2.0 specification for no-load power is less stringent for AC/AC EPSs and
because AC/AC EPSs do not have overhead circuitry to allow them to reduce power
consumption in no-load mode, as discussed in section 2.3.2.2. Therefore, DOE applied the new
equations to product classes A2, A3, and A4 according to how they related to A1, as shown in
Table 2.8.
Table 2.21 presents the low-voltage efficiency CSLs (VLOW), and Table 2.22 presents the
AC/AC no-load power CSLs (PNL_AC). Both the low-voltage efficiency and AC/AC no-load
power CSLs were created by scaling based on the CSLs from product class A1 and ENERGY
STAR 2.0, which has the corresponding four-product class structure. See chapter 5 for more
details on how DOE created the low-voltage efficiency and AC/AC no-load power CSLs.

2-46

Table 2.21 Active-Mode Efficiency CSLs for Low-Voltage EPSs (Product Classes A2 and
A4)
CSL Nameplate Output Power (Pno) Minimum Average Efficiency in Active Mode
(expressed as a decimal)
CSL 0 Identical to Basic Voltage CSL 0
Identical to Basic Voltage CSL 0
0 to 1 watt
CSL 1

0.497 Pout + 0.067

> 1 to 49 watts

0.075 ln (Pout) + 0.561

> 49 watts

0.860

0 to 1 watt
CSL 2

0.507 Pout + 0.077

> 1 to 49 watts

0.0825 ln (Pout) - 0.0011 Pout + 0.586

> 49 watts

0.860

0 to 1 watt
CSL 3

0.517 Pout + 0.087

> 1 to 49 watts

0.0834 ln (Pout) - 0.0014 Pout + 0.609

> 49 watts

0.870

0 to 1 watt
CSL 4

0.537 Pout + 0.097

> 1 to 49 watts

0.0914 ln (Pout) - 0.0014 Pout + 0.609

> 49 watts

0.900

Table 2.22 No-load Power CSLs for AC/AC EPSs (Product Classes A3 and A4)
CSL Nameplate Output Power (Pno) Maximum Power in No-Load
Ac-Ac EPS
CSL 0
0 to 250watt
0.500 watts
CSL 1

0 to 250watt

0.500 watts

CSL 2

0 to 250watt

0.230 watts

CSL 3

0 to 250watt

0.210 watts

CSL 4

0 to 250watt

0.165 watts

2-47

2.5.3 Engineering Analysis for Non-Class A EPSs


DOE developed its non-Class A EPS engineering analysis using a combination of
approaches. In particular, DOE used the approach from its non-Class A EPS determination
analysis and the Class A EPS analysis developed for this preliminary analysis. DOE used this
approach for multiple-voltage and high-power EPSs because DOE developed the analysis using
the same methods of testing and teardowns and manufacturer interviews as used for Class A
EPSs in this preliminary analysis so the determination analysis is consistent with the current
analytical approach. DOE notes that interested parties did not question this approach when
commenting on the draft determination analysis that DOE had prepared. Similarly, DOE based
its analysis for medical and MADB EPSs on the approach contained in the preliminary analysis
for Class A EPSs because DOE is currently unaware of any technical limitations for medical and
MADB EPSs to achieve the same efficiencies at the same costs as Class A EPSs. DOE also notes
that interested parties raised no questions regarding this approach, except in the context of
MADB wall adapters equipped with charge control. However, the presence of charge control
indicates that an MADB wall adapter is not an EPS and, therefore, would not be within the scope
of this rulemaking.
2.5.3.1 Non-Class A EPS Representative Product Classes and Representative Units
In this section DOE presents its selection of non-Class A EPS representative product
classes and representative units based on the non-Class A EPS product classes presented in
chapter 3. As noted previously, DOE based the non-Class A EPS product classes on the product
classes from the non-Class A EPS determination analysis as well as the product classes
developed in this preliminary analysis for Class A EPSs.
Representative Product Classes and Units for Multiple-Voltage EPSs. DOE chose to
treat both multiple-voltage EPS product classes as representative because there are significant
differences between the two classes as shown in Table 2.23. As explained in detail in the
determination analysis, representative product class X1 consists mainly of EPSs for imaging
equipment. These multiple-voltage EPS have similar nameplate output power ratings for each
output voltage. In contrast, EPSs in product class X2 are for video game applications and
typically have one output that provides the majority of the output power. Because a single
application constitutes the majority of shipments in each representative product class, DOE
decided to use those applications as representative units as shown in Table 2.24.
Table 2.23 Multiple Voltage EPS Representative Product Classes
Nameplate Output Power
Product Class
< 100 watts
X1 (representative
100 watts
X2 (representative)
Table 2.24 Multiple-Voltage EPS Representative Units
Nameplate Output Nameplate Output Second Nameplate Example Application
Power [W]
Voltage [V]
Output Voltage [V]
40
16
32
Multi-Function Device
203
5
12
Video Game

2-48

Representative Product Classes and Units for High-Power EPSs. DOE chose to treat
the high-power EPS product class as representative because it is the only product class (shown in
Table 2.25). As explained in detail in the determination analysis, representative product class H1
consists mainly of EPSs for ham radios. Because a single application constitutes the majority of
shipments in the representative product class, DOE decided to use that application to define the
attributes of the representative unit shown in Table 2.26.
Table 2.25 High Power EPS Representative Product Classes
Nameplate Output Power Product Class
> 250 watts
H1 (representative)
Table 2.26 High-Power EPS Representative Units
Nameplate Output Power [W] Nameplate Output Voltage [V] Example Application
345
13.8
Amateur Radio
Representative Product Classes and Units for Medical EPSs. DOE chose to treat only
medical EPS product class M1 as representative because there are significant similarities
between the product classes shown in Table 2.27. Specifically, the medical EPS product classes
have the same product class structure as the Class A EPS product classes, as explained in chapter
3. To that end, DOE applied the same structure for the medical EPS representative product
classes and representative units (shown in Table 2.28).
DOE took this approach because it is consistent with the approach for medical EPSs in
the determination analysis. Specifically, the determination analysis presented representative
product classes and units for medical EPSs based on Class A EPSs. Interested parties did not
question the validity of this approach in the determination analysis. Since publishing the
determination analysis, DOE has enhanced its understanding of Class A EPSs through additional
testing, teardowns, and manufacturer interviews. Consequently, the representative product
classes and units presented in Table 2.27 and Table 2.28 are consistent with the Class A EPS
analysis in chapter 5.
Table 2.27 Proposed Medical EPS Representative Product Classes
Basic Voltage Output Low Voltage Output*
AC-DC Conversion
M1 (representative)
M2 (scaled)
AC-AC Conversion
M3 (scaled)
M4 (scaled)
* Low voltage output EPSs have nameplate output voltage less than six volts and nameplate output current greater
than or equal to 550 milliamps. All other EPSs are basic voltage output.

Table 2.28 Possible Medical EPS Representative Units


Representative Nameplate Output
Nameplate Output
Unit
Power
Voltage
[watts]
[volts]
1
2.5
5
2
18
12
3
60
15
4
120
19

2-49

Example
Application
Blood Pressure Monitor
Medical Nebulizer
Sleep Apnea Machine
Portable Oxygen Concentrator

Representative Product Classes and Units for MADB EPSs. DOE chose to treat only
MADB EPS product class B1 as representative because there are significant similarities between
the product classes as shown in Table 2.29. Specifically, the MADB EPS product classes have
the same product class structure as the Class A EPS product classes, as explained in chapter 3.
To that end, DOE applied the same structure for the MADB EPS representative product classes
and representative units (shown in Table 2.30). Since DOE can apply all Class A EPS analysis to
MADB EPSs, it intends to include a 120-watt representative unit, if necessary. However, Table
2.30 reflects that DOE has not identified an associated application for this unit. As noted in the
executive summary, DOE is seeking stakeholder comment on any applications that use a 120
watt MADB EPS and whether to analyze this unit if it cannot identify any associated
applications.
DOE took this approach because it is consistent with the approach for MADB EPSs in
the determination analysis. Specifically, in the determination analysis DOE created
representative product classes and units for MADB EPSs based on Class A EPSs. Interested
parties did not question the validity of this approach in the determination analysis. As with
medical EPSs, since publishing the determination analysis, DOE has further enhanced its
understanding of Class A EPSs through additional testing, teardowns, and manufacturer
interviews. Consequently, the representative product classes and units presented in Table 2.29
and Table 2.30 are consistent with the Class A EPS analysis in chapter 5.
Table 2.29 Proposed MADB EPS Representative Product Classes
Basic Voltage Output Low Voltage Output*
AC-DC Conversion
AC-AC Conversion

B1 (representative)
B3 (scaled)

B2 (scaled)
B4 (scaled)

* Low voltage output EPSs have nameplate output voltage less than six volts and nameplate output current greater
than or equal to 550 milliamps. All other EPSs are basic voltage output.

Table 2.30 Possible MADB EPS Representative Units


Representative Nameplate Output
Nameplate Output Voltage
Unit
Power
[volts]
[watts]
1
2.5
5
2
18
12
3
60
15
4
120
19

Example
Application
RC Toy
DIY Power Tool
Motorized Bicycle
[none identified]

2.5.3.2 Non-Class A EPS Candidate Standard Levels of Efficiency


In this section, DOE presents its selection of CSLs for non-Class A EPSs and explains
how it will develop the CSL equations in the NOPR. DOE based these levels using the CSLs
from the determination analysis and the preliminary analysis for Class A EPSs.
Candidate Standard Levels of Efficiency for Multiple-Voltage EPSs. As indicated
earlier, DOE based its development of CSLs for multiple-voltage EPSs on those presented in the

2-50

non-Class A EPS determination analysis. DOE took this approach primarily because multiplevoltage EPSs are distinct from single-voltage EPSs in two ways. First, they share different
underlying technologies, particularly, different overhead circuitry to control power flow to each
output. This factor is significant because these circuit differences may result in a different range
of efficiency levels, and possibly a max tech level that is lower than the single-voltage EPS level,
for the same output power. Second, there are currently no established standard levels in the
multiple-voltage EPS market. Consequently, DOE structured the CSLs for multiple-voltage EPSs
based on products available in the market and the theoretical maximum technologically feasible
level described by manufacturers as shown in Table 2.31.
Table 2.31 Multiple-Voltage EPS Candidate Standard Levels of Efficiency
Number Reference
Basis
CSL 0
Market Bottom Least efficient units in the market
CSL 1
Mid Market
Efficiency in the middle of the market
CSL 2
Best in Market Curve fit to most efficient test unit data points
CSL 3
Max Tech
Curve fit to manufacturer max tech data points
In the determination analysis, DOE based its efficiency values for each CSL on test and
manufacturer data. For the NOPR, DOE may gather more data for multiple-voltage EPSs with
which to develop specific efficiency values for each CSL. In contrast to Class A EPSs whose
minimum average efficiency and maximum no-load power consumption requirements are
functions of nameplate output power, DOE plans to use a single efficiency metric for each CSL.
DOE considers this approach to be appropriate because although product classes X1 and X2 span
a range of nameplate output powers, DOE is aware of only one significant application in each
product class, requiring only one efficiency value. Further, because DOE is currently unaware of
other significant applications, DOE is unable to characterize other products in the market to
develop an equation similar to the CSL equation used for single-voltage EPSs.
Candidate Standard Levels of Efficiency for High-Power EPSs. DOE based its
development of CSLs for high-power EPSs on those presented in the non-Class A EPS
determination analysis. Specifically, high-power EPSs are distinct from other types of EPSs for
two reasons: their underlying technology is different and there are no established standard levels
in the high-power EPS market. Consequently, DOE structured the CSLs for high-power EPSs
based on products available in the market and the theoretical maximum technologically feasible
level described by manufacturers, as shown in Table 2.32.
Table 2.32 High-Power EPS Candidate Standard Levels of Efficiency
Number Reference
Basis
CSL 0
Market Bottom Least efficient units in the market
CSL 1
Mid Market
Efficiency in the middle of the market
CSL 2
Best in Market Curve fit to most efficient test unit data points
CSL 3
Max Tech
Curve fit to manufacturer max tech data points
The determination analysis used efficiency values for each CSL on test data and
manufacturer data. For the NOPR, DOE may gather more data for high-power EPSs with which
to develop specific efficiency values for each CSL. In contrast to Class A EPSs whose minimum

2-51

average efficiency and maximum no-load power consumption requirements are functions of
nameplate output power, DOE intends to have a single efficiency metric for each CSL. DOE
believes this approach is appropriate because although product class H1 spans a range of
nameplate output powers, it has only one significant application and therefore only needs one
value.
Candidate Standard Levels of Efficiency for Medical EPSs. DOE based its CSLs for
medical EPSs using the levels developed for Class A EPSs (product class A1). DOE adopted
this approach because medical EPSs share two key similarities. First, their underlying
technology is similar, which indicates that they can achieve the same efficiencies. Second,
because medical EPSs and Class A EPSs similar technologies, the established standards for Class
A EPSs can be applied to medical EPSs as well.
In spite of these similarities, however, there are also two key differences between medical
EPSs and Class A EPSs. First, medical EPSs are exempt from the Class A energy efficiency
requirements set out by EISA 2007. In light of the fact that medical EPSs are not covered by the
Class A standards, DOE currently believes that medical EPSs may be less efficient than their
Class A EPS-counterparts, although DOE did not find any medical EPSs with efficiency levels
less than the EISA level in the determination analysis. This possibility led DOE to the tentative
view that there may be medical EPSs less efficient than EISA standards and ultimately
influenced the agencys range of potential CSLs from which to choose. Second, as noted in
chapter 3, medical EPSs conform to additional safety requirements. These requirements make the
absolute price of medical EPSs higher than Class A EPSs, but do not affect medical EPS
efficiency or incremental cost. Consequently, DOE structured the CSLs for medical EPSs based
on the Class A EPS CSLs and set the baseline CSL 0 to the least efficient Class A EPS unit used
to characterize the medical EPS market as shown in Table 2.33. All other CSL equations come
directly from the Class A EPS analysis. For example, the baseline CSL 0 for Class A EPSs is
CSL 1 for medical EPSs.
Table 2.33 Medical EPS Candidate Standard Levels of Efficiency
Number Reference
Basis
CSL 0
Market Bottom
Least efficient units in the market
CSL 1
EISA 2007
EISA 2007 equations for efficiency and no-load power
CSL 2
ENERGY STAR 2.0 ENERGY STAR 2.0 equations for efficiency and no-load
power
CSL 3
Intermediate
Curve fit to data points
CSL 4
Best in Market
Curve fit to most efficient test unit data points
CSL 5
Max Tech
Curve fit to manufacturer max tech data points
In the determination analysis DOE based its efficiency values for each CSL on existing
levels of efficiency in the market, test data, and manufacturer data. For the NOPR, DOE is
considering gathering more data for medical EPSs with which to develop specific efficiency
values for the baseline CSL 0. This approach would enable DOE to characterize the benefits of
higher standards to consumers purchasing EPSs with efficiencies below EISA requirements.
However, DOE believes that those consumers constitute a small and shrinking part of the market,
based on information from manufacturer interviews. Therefore, DOE is also considering

2-52

accounting for the lowest part of the market in the EISA qualified CSL as a simplifying
alternative. Under this alternate approach, there would be no difference in treatment between
Class A EPSs and medical EPSs in the engineering analysis. As noted in the executive summary,
DOE is seeking stakeholder comment on which approach to use.
Regardless of how DOE treats EPSs that are not covered by the EISA 2007-prescribed
standards, DOE intends to apply data from Class A EPSs directly to medical EPSs for those
CSLs that are common between the two products. For each CSL, DOE intends to have equations
that are functions of nameplate output power similar to Class A EPSs. In particular, DOE intends
to apply the Class A EPS equations directly to the corresponding medical EPS CSLs. In addition,
to scale medical EPS CSL equations from product class M1 to the other product classes, DOE
intends to use the Class A EPS methodology and results. DOE believes this approach is
appropriate because of the similarity between medical EPSs and Class A EPSs.
Candidate Standard Levels of Efficiency for MADB EPSs. DOE developed its CSLs for
MADB EPSs using the CSLs for Class A EPSs (product class A1) presented in chapter 5.
Specifically, MADB EPSs are similar to Class A EPSs in two ways: their underlying technology
is similar, and there are established standard levels that can be applied to the MADB EPS
market. However, there is one difference between MADB EPSs and Class A EPSs, which
affected DOEs choice of CSLs. Namely, MADB EPSs are not covered by the EISA 2007
standards, which led DOE to believe that some MADB EPSs may be less efficient than Class A
EPSs because they are not required to meet EISA standards. In contrast to medical EPSs, DOE
believes that MADB EPSs do not conform to additional requirements, safety or otherwise.
Consequently, DOE structured the CSLs for MADB EPSs based on the Class A EPS CSLs and
set the baseline CSL 0 to the least efficient MADB EPS units in the market as shown in Table
2.34. All other CSL equations come directly from the Class A EPS analysis. For example, the
baseline CSL 0 for Class A EPSs is CSL 1 for MADB EPSs.
Table 2.34 MADB EPS Candidate Standard Levels of Efficiency
Number Reference
Basis
CSL 0
Market Bottom
Least efficient units in the market
CSL 1
EISA 2007
EISA 2007 equations for efficiency and no-load power
CSL 2
ENERGY STAR 2.0
ENERGY STAR 2.0 equations for efficiency and noload power
CSL 3
Intermediate
Curve fit to data points
CSL 4
Best in Market
Curve fit to most efficient test unit data points
CSL 5
Max Tech
Curve fit to manufacturer max tech data points
In the determination analysis DOE based its definitions of the efficiency values for each
CSL on existing levels of efficiency in the market, test data and manufacturer data. For the
NOPR, DOE is considering gathering more data for MADB EPSs with which to develop specific
efficiency values for the baseline CSL 0. This would enable DOE to characterize the benefits of
higher standards to consumers purchasing EPSs with efficiencies below EISA requirements.
However, DOE believes that those consumers constitute a small and shrinking part of the market,
based on information from manufacturer interviews. Therefore, DOE is also considering
accounting for the lowest part of the market in the EISA qualified CSL as a simplifying

2-53

alternative. Under this alternate approach, there would be no difference between Class A EPSs
and MADB EPSs in the engineering analysis. As noted in the executive summary, DOE is
seeking stakeholder comment on which approach to pursue.
Regardless of how DOE treats EPSs that are not covered by the EISA 2007 standards,
DOE intends to apply data from Class A EPSs directly to MADB EPSs for those CSLs that are
common between the two products. For each CSL, DOE intends to have equations that are
functions of nameplate output power similar to Class A EPSs. In particular, DOE intends to
apply the Class A EPS equations directly to the corresponding MADB EPS CSLs. In addition, to
scale MADB EPS CSL equations from product class M1 to the other product classes DOE
intends to use the Class A EPS methodology and results. DOE believes this approach is
appropriate because of the similarity between MADB EPSs and Class A EPSs.
2.5.3.3 Developing Cost-Efficiency Results for non-Class A EPSs
In this section DOE presents its methodology to develop cost-efficiency curves for nonClass A EPSs. DOE intends to develop cost-efficiency curves based on the data used to generate
cost-efficiency curves in the determination analysis as well as the data used to generate costefficiency curves for Class A EPSs in this preliminary analysis.
To develop the NOPR, DOE intends to use the same scaling and analysis techniques for
both Class A EPSs and non-Class A EPSs. Thus, DOE intends to revise the data from the nonClass A EPS determination analysis for the NOPR. Specifically, in developing the Class A EPS
engineering analysis in this preliminary analysis, DOE refined the scaling methodologies it used
in the non-Class A EPS engineering analysis in the determination analysis. In addition, DOE
evaluated EPSs based on the full costs of all materials, as opposed to the determination analysis,
which evaluated EPSs based on the efficiency-related materials cost. DOE believes that its
refined scaling methods have enhanced the robustness of the analysis, and that the full cost of an
EPS is a more appropriate basis for evaluation because EPS components are interrelated. The
following subsections explain how DOE plans to use data from the determination analysis; in all
cases, DOE would apply the updated scaling and analysis techniques to the data.
Cost-Efficiency Curves for Multiple-Voltage EPSs. DOE developed its cost-efficiency
curves for multiple-voltage EPSs using data obtained by testing and tearing down EPSs and from
manufacturer interviews. DOE intends to use a similar approach in the NOPR and believes that
the cost-efficiency curves for multiple-voltage EPSs that it presented in the determination
analysis are reasonably accurate because it used the same analytical methods as were used for
Class A EPSs in this preliminary analysis. DOE also notes that it received no comments from
interested parties pointing to flaws in its approach. Therefore, DOE intends to use that data (with
updated scaling and analysis techniques) to develop the cost-efficiency curves in the NOPR for
product classes X1 and X2.
Cost-Efficiency Curves for High-Power EPSs. The determination analysis also
presented cost-efficiency curves for high-power EPSs that relied on data obtained by testing and
tearing down EPSs, as well as interviewing manufacturers. DOE intends to use a similar
approach in the NOPR. Specifically, DOE believes that the cost-efficiency curves for high-power
2-54

EPSs that it presented are reasonably accurate because it used the same analytical methods as
were used for Class A EPSs in this preliminary analysis. . DOE also received no comments
pointing to flaws with the use of this approach. Therefore, DOE intends to use that data (with
updated scaling and analysis techniques) to develop the cost-efficiency curves in the NOPR for
product class B1.
Cost-Efficiency Curves for Medical EPSs. To develop cost-efficiency curves for product
class M1 for medical EPSs in the NOPR, DOE intends to apply the cost-efficiency curves for
Class A EPSs because it believes medical EPSs to be technically equivalent to Class A EPSs
other than that they conform to more stringent safety standards, which does not affect their
incremental cost and therefore does not affect their cost-efficiency relationship. . In addition,
DOE may conduct testing, teardowns, and manufacturer interviews to gather data on EPSs that
are not covered by the EISA 2007-prescribed standards.
Cost-Efficiency Curves for MADB EPSs. To develop cost-efficiency curves for product
class B1 for MADB EPSs in the NOPR DOE intends to apply the cost-efficiency curves for
Class A EPSs because it believes MADB EPSs to be technically equivalent to Class A EPSs such
that they share the same cost-efficiency relationship. In addition, DOE will conduct testing,
teardowns, and manufacturer interviews to gather data on EPSs with efficiencies less than the
EISA standard.
2.5.4 Engineering Analysis for Battery Chargers
2.5.4.1 BC Representative Product Classes and Representative Units
As discussed earlier in section 2.3.2, the preliminary analysis divides BCs into 10 product
classes. DOE adopted this approach after analyzing comments from interested parties and
examining market-available BC technologies.
Further examination of BC units in the market led to a subset of the 10 product classes
being selected for further analysis as the representative product classes. These classes were
analyzed through testing, teardowns, and interviews with manufacturers. Collectively, the classes
constitute the majority of BC shipments and national energy consumption. By focusing its
analysis on the representative product classes, DOE can provide the most accurate standards
analysis for the applications that have the largest impact on users (due to high shipments or high
per-unit energy consumption) or that have the largest impact on the nation (due to high
shipment-weighted energy consumption). Table 2.35 presents the representative product classes
in the context of the BC product classes developed in chapter 3.

2-55

Table 2.35 BC Representative Product Classes


Product
Class
Number Product Class
Inductive
Low Energy,
1
Connection
Inductive
Low Energy,
<
<4 V
2
Low Voltage
100
Low Energy,
Wh
410 V
3
AC
Med. Voltage
In,
Low Energy,
>10 V
4
DC
High Voltage
Out
Med. Energy,
<20 V
5
100
Low Voltage
3000
Med. Energy,
Wh
20 V
6
High Voltage
> 3000 Wh

<9 V Input

9 V Input

AC Output
from
Battery

10

DC In,
DC Out

AC In,
AC Out

High Energy
Low Energy,
Low Voltage
DC Input
Low Energy,
High Voltage
DC Input
Low Energy,
AC Output

Representative or Scaled Product


Class
Representative
Representative
Scaled
Representative
Representative
Scaled
Representative
Scaled

Scaled

Representative

DOE subsequently focused its analysis on one representative unit, an idealized BC


typical of those used for high-volume applications in a given representative product class.
Because results from the analysis of these representative units would later be extended to
additional BCs, they were selected from high-volume and/or high-energy-consumption
applications that use batteries that are typical across BCs in the product class.
For each representative product class, DOE then identified the most common battery
voltage and energy combinations from a survey of popular battery-operated products in the
market, taking into account the distribution of those BC characteristics by application. DOE then
selected the BC characteristics of the representative units to correspond to the densest clusters of
BC models. Interested parties did not comment on the selection of representative units for
analysis, except for Delta-Q, which commented that DOE should consider testing its QuiQ
charger because it is representative of the 1000 watt class of BCs. (Delta-Q, No. 15 at p. 1)
DOE purchased and tested several BCs for golf cars with output power between 850 and
1000 watts, intended for charging a 48 volt, 3750 watt-hour flooded lead-acid battery. This
representative unit, and the other representative units, are presented in Table 2.36.

2-56

Table 2.36 The BC Representative Units for each Representative Product Class
Avg. Annual
Product
Rep. Unit
Rep. Unit
Production
Class
Voltage
Energy
Volume
Number
V
Wh
K units
Inductive
1
3.6
1.5
500
Connection
480
<4 V
2
3.6
3
< 100
AC
In,
DC
Out

Wh

100
3000
Wh

410 V

>10 V

10.8

20

640

<20 V

12

800

50

20 V

> 3000 Wh

Scaled Product Class

Scaled Product Class

48

3,750

DC In,
DC Out

<9 V Input

Scaled Product Class

9 V Input

Scaled Product Class

AC In,
AC Out

AC Output
from Battery

10

12

70

150

1000

2.5.4.2 BC Candidate Standard Levels Analyzed


In its engineering analysis of BCs, DOE examined the cost impacts of improving the
efficiency of each of the representative units presented in Table 2.36 to evaluate potential energy
efficiency standards. DOE performed this analysis by evaluating the design options and
corresponding costs necessary to increase the efficiency of currently-available BCs to each
candidate standard level (CSL).
CSLs are generally based on (1) design options associated with the specific units being
analyzed; (2) other voluntary specifications or mandatory standards that cause manufacturers to
develop products at particular efficiency levels; and (3) the maximum technologically feasible
level.h For example, in the EPS case, DOE considered the current DOE standard level as the
baseline CSL and the ENERGY STAR level as another, higher CSL (see section 2.5.2.2).
Several interested parties commented on the CSLs that DOE should use for the BC
preliminary analysis. PG&E commented that DOE should include CSLs that encompass design
options that are not currently common in the marketplace. (PG&E, No. 20 at p. 12) Similarly,
PG&E recommended that DOE look at a wide variety of efficiencies across different BCs.
(PG&E, Pub. Mtg. Tr., No. 14 at p. 58)
h

The max-tech level represents the most efficient design that is commercialized or has been demonstrated in a
prototype with materials or technologies available today. Max tech is not constrained by economic justification
apart from the requirement that there be more than one unique way to achieve it, per the discussion in 2.5.2.2. It is
typically the most expensive design option considered in the engineering analysis.

2-57

The max-tech CSL mentioned above is based on technologies that are typically only
available in prototype form and not commercially available. Furthermore, DOE also intends to
analyze the best-in-market level for the preliminary analysis. Because this level will be based on
annual energy consumption of the most efficient commercially available BCs, DOE does not
expect their design options to be common, as the majority of the market will have lower
efficiency. DOE also intends to analyze these lower efficiency levels, including the least efficient
BCs for each representative unit battery voltage and energy, satisfying PG&Es request to
evaluate a wide variety of BCs.
CEC et al. requested that DOE analyze the efficiency standard for BCs that PG&E
proposed to the State of California to adopt as part of its efficiency standards -- PG&E, in turn,
asked that DOE adopt that proposed standard. (CEC et al., No. 19 at p. 1; PG&E, Pub. Mtg. Tr.,
No. 14 at p. 54) That standard, if adopted, would specify a specific 24-hour energy efficiency
that is only dependent upon a battery chargers rated battery energy. As explained above, the
evaluation of BCs currently in the market underlies DOEs engineering analysis except for the
max-tech CSL, which cannot be met by any commercially available BCs and must be derived
and evaluated through modeling and interviews with manufacturers. According to PG&Es data,
which were presented to DOE at the framework document public meeting, there are few BCs for
batteries with energies above 10 watt-hours that can meet PG&Es proposed standard. (PG&E
No. 13 at pp. 79)
Although DOEs testing identified additional BCs in the market that could meet the
proposed PG&E standards (including ones with battery energies greater than 10 watt-hours), the
BCs energy consumption in each mode does not always fall near the PG&E proposed standards
level, sometimes exceeding the standard by a wide margin. Therefore, a direct evaluation of the
PG&E standards proposal may not always be possible, even though some of the best-in-market
and all the max-tech units analyzed would meet the proposed standard.
However, DOE intends to examine a variety of applications in the market. In addition to
the baseline, best-in-market, and max-tech levels, DOE also analyzed BC units at an improved
level, where available. This level is not associated with any standard or guideline, but rather
reflects the energy consumption achievable by units in the market through the application of
some of the energy efficiency design options presented earlier (see section 2.3.3.4).
These CSLs are summarized for each representative unit in Table 2.37 below, expressed
in terms of annual energy consumption, calculated by summing the BC energy consumption in
each mode, assuming a class-average usage profile.

2-58

Table 2.37 Candidate Standard Levels for the BC Representative Units


Annual Energy Consumption by Representative Unit
kWh
Low
Low
Med.
Low
Low
Energy,
Energy, Energy,
Energy,
Energy,
Low
High
Low
High
AC
Inductive Voltage
Voltage Voltage
Energy
Output
1.5 Wh,
3.6 V

3 Wh,
3.6 V

20 Wh,
10.8 V

800 Wh,
12 V

3750 Wh,
48 V

70 Wh
12 V

0 Baseline

10.0

10.0

39.4

202.7

290.0

19.6

1 Improved

7.0

6.0

10.5

159.6

250.0

6.4

Best-in-Market

3.5

1.2

6.1

100.0

200.0

4.0

Maximum
3 Technologically
Feasible

1.5

75.0

150.0

1.5

# Efficiency Level

2.5.4.3 BC Teardown Analysis


As mentioned in the discussion above, the CSLs used in the BC engineering analysis
were based on the efficiencies of BCs available in the market. Following testing, the units
corresponding to each commercially available CSL were disassembled to (1) evaluate the
presence of energy efficient design options and (2) torn down to estimate material costs. The
disassemblies were performed by one of DOEs subject-matter experts and included an
examination of the general design of the BC and an evaluation of the presence of any of the
technology options discussed in chapter 3.
After the BC units corresponding to the CSLs were evaluated, they were torn down by
iSuppli, another DOE contractor. These teardowns were comparable to those conducted for EPSs
and described in section 2.5.2.3, with the exception of BCs embedded inside complex consumer
electronic applications such as camcorders and notebook computers.
Because the BC constitutes a small portion of the circuitry of these products, DOE,
through iSuppli (a firm specializing in consumer electronics costs), identified the subset of
components in each product enclosure responsible for battery charging, including the battery,
charge regulator, and any related power converters and voltage regulators. The function of the
latter two subcircuits was split between the battery charger and other aspects of the application
(e.g., powering the notebook computer in addition to charging its battery). Nonetheless, because
of the crucial role played by these subcircuits in the battery charging process, their full cost was

2-59

included in the BC manufacturing cost estimate. Details regarding BC teardowns for the
representative units are in chapter 5.
2.5.4.4 BC Manufacturer Interviews
Navigant Consulting, a DOE contractor, entered into non-disclosure agreements and
interviewed representatives of several firms that manufacture battery-powered products or BCs
for those products. Navigant Consulting aggregated the responses from these interviews and
presented DOE with generalized responses free of any proprietary data for use in the analysis.
During the course of each interview, interviewers asked the representatives to provide
information about technology options for increasing the energy efficiency of BCs that ship with
their products. Prior to each interview, each manufacturer received a Cost-Efficiency Estimation
Survey (included as appendix 5C) to guide their responses. To ensure consistency between
manufacturers, the survey specified the parameters of each BC representative unit under
consideration. See Table 2.36.
For each representative unit, the interviewers asked manufacturers to describe the
technological improvements and associated costs necessary to meet each of the CSLs presented
in section Table 2.37. These CSLs were presented in a disaggregated form (i.e., energy
consumption by mode, rather than combined into a weighted annual energy consumption) to help
the respondents. Finally, manufacturers were also encouraged to present data on BCs they were
most familiar with, even if those did not perfectly match with the CSLs listed in Table 2.35. An
aggregated summary of their responses, divided by representative unit, is presented in chapter 5.
2.5.4.5 BC Cost Model
Similar to EPSs, DOE gathered inputs on markups for BCs from manufacturer
interviews. In particular, manufacturers were asked to provide typical markups (due to
production and non-production costs, as well as profits) of BC suppliers, OEMs, and distributors
and retailers. These markups were subsequently used to adjust the price of a BC at any point in
the supply chain, whether that point was identified through manufacturer interviews or
teardowns, to the manufacturer selling price (MSP), charged by the BC supplier to the OEM.
In either case, the analysis focused on the cost of BC components directly related to
efficiencyi.e., the electronics. The analysis excluded the cost of the packaging, output cord,
and cosmetic touches. These may vary from product to product, and their costs depend greatly on
the application and the degree of integration of the BC with the end-use product. The markups
for each product class are presented in detail in chapter 5.
2.5.4.6 BC Relationships between Cost and Efficiency
DOE integrated the results of the BC teardown analysis with the information obtained
during manufacturer interviews, including markups, to arrive at the relationship between cost and
efficiency for the BC representative units. This cost of meeting each candidate standard level of
efficiency is the main result of the engineering analysis and is used in both the life-cycle cost
(LCC) and payback period (PBP) analysis and the national impacts analysis (NIA).

2-60

Following the publication of the framework document, Micro Power Electronics (MPE)
noted that setting an efficiency standard above 80% could exclude the single-ended primary
inductor converter (SEPIC) architecture, which is low cost. (MPE, No. 5 at p. 1) This comment
highlights some of the issues considered during the engineering analysis, and one of its expected
outcomesthat the cost of BCs is likely to increase for higher CSLs. The full engineering results
for each representative unit, including the designs necessary to meet each CSL, are presented in
chapter 5.
2.5.4.7 BC Scaling from Representative Units to Product Classes
Following the development of engineering results for the representative units, DOE must
extend this analysis to all BCs that were not analyzed but covered by the standard. This task is
twofold: (1) scaling the representative unit results to BCs that are also in the representative
product class, but which differ in battery voltage and energy; and (2) scaling the representative
unit results to BCs in product classes not explicitly analyzed.
As mentioned in section 2.3.2.1, PG&E suggested that DOE mitigate boundary issues by
using a continuous function when setting the standard. (PG&E, No. 20 at p. 7) Such a function
would also be used to scale the engineering results from the representative units to other BCs not
explicitly analyzed.
However, developing the correct scaling relationships presents a challenge because of the
wide variety of BC designs, each of which responds differently to changes in battery voltage and
energy. For example, PG&E commented that DOE cannot use scaling relationships for AC-AC
power supplies when developing scaling relationships for BCs that use switched-mode AC-DC
power supplies. (PG&E, No. 20 at p. 13) Similarly, Delta-Q commented that DOE should not
scale BC engineering analysis findings across large differences in output power, and should
consider the applicability of new technologies on a case-by-case basis. (Delta-Q, No. 15 at p. 1)
However, neither commenter provided a basis for their recommendations.
In response to these comments, DOE used a hybrid approach to scaling in the preliminary
analysis. When possible, DOE first attempted to use engineering relationships to scale the
performance of BC designs at each CSL, dividing the problem by operational mode and BC
stage (i.e., power supply, charge controller), such that the final energy consumption of a scaled
unit was a function of the expected performance of each stage in each mode, weighted by the
average time spent in each mode.
In cases where this was not possible, DOE scaled the results based on test results of
actual BCs available in the market. Overall, DOE attempted to minimize the effects of scaling
BC performance by directly analyzing a number of representative units across a broad range of
BC output power.
For the preliminary analysis, DOE used the engineering-based method to obtain costefficiency results for the scaled product classes. By modeling the energy consumption of the
representative units in situations typical of BCs in other classes (summarized in Table 2.38),
DOE was able to estimate the cost and efficiency of meeting the CSLs as shown in Table 2.39.

2-61

DOE has not yet scaled the CSLs for each representative unit (and the scaled units in Table 2.38)
to all the remaining BCs within each product class.
Table 2.38 The BC Representative Units for each Representative and Scaled Product Class.
Avg. Annual
Product
Rep. Unit
Rep. Unit
Production
Class
Voltage
Energy
Volume
Number
V
Wh
K units*
Inductive
1
3.6
1.5
500
Connection
480
<4 V
2
3.6
3
< 100
AC
In,
DC
Out

Wh

410 V

7.2

10

480

>10 V

10.8

20

640

<20 V

12

800

50

20 V

36

384

48

3,750

50
150

100
3000
Wh

> 3000 Wh
DC In,
DC Out

<9 V Input

3.6

480

9 V Input

3.6

480

AC In,
AC Out

AC Output
from Battery

10

12

70

1000

* Note: The production volume for product classes 3, 8, 9 uses product class 2 data, and the production volume for
product class 6 uses product class 5 data.

Table 2.39 Candidate Standard Levels for the Scaled BC Representative Units.
Annual Energy Consumption by Representative Unit
kWh
Low Energy,
Medium
Low Energy,
Low Energy,
Medium
Energy, High
Low Voltage
High Voltage
Voltage
Voltage
DC Input
DC Input
#
0

Efficiency Level
Baseline

10Wh, 7.2V
10.0

384Wh, 36V
69.3

2Wh, 3.6V
1.5

5Wh, 3.6V
1.3

Improved

5.4

41.9

0.9

0.8

Best-in-Market

1.0

25.0

0.8

0.5

Maximum
Technologically
Feasible

18.0

2-62

2.6

MARKUPS TO DETERMINE PRODUCT PRICE

DOE used markups to convert the manufacturer selling prices (MSPs) of BCs and EPSs,
which were calculated in the engineering analysis, to consumer prices, which then were used in
the life-cycle cost (LCC) and payback period (PBP) and national impact analyses. DOE
calculates markups for baseline products (baseline markups) and for more efficient products
(incremental markups). The incremental markup relates the change in the MSP of higherefficiency models (the incremental cost increase) to the change in the retailer or distributor sales
price.
DOE made the following key assumptions in calculating markups:
The dominant path to market establishes the retail price and, thus, the composite markup
for a product.
The markups applied to end-use products that use BCs and EPSs are approximations of
BC and EPS markups.
The baseline markups that manufacturers and retailers apply to end-use products that use
BCs and EPSs are equal to those companies average markups across their entire product
lines.
Expenses like labor and administrative costs remain fixed and need not be recovered in
the incremental markup. Profits and other operating costs are assumed to be variable and
to scale with the MSP.
Markups can be derived from inspection of companies public financial filings.
To determine which markups are applied to the MSP, DOE first modeled the distribution
channels for BCs and EPSs. Research indicated that the most common path begins with BC/EPS
manufacturers, progresses through end-use product manufacturers (OEMs), and ends with
consumer product retailers or distributors (retailers), at which point a sales tax is applied and the
BC or EPS is sold to consumers. DOE relied on publicly available corporate financial data to
estimate individual product markups.
Vivitar and Earthjustice commented that the increase in manufacturing costs from
producing more efficient units is not always passed directly to consumers. Rather, some
manufacturers may reduce markups to offset these increased costs. (Vivitar, Pub. Mtg. Tr., No.
14 at p. 242; Earthjustice, No. 18 at p. 2) DOE recognizes that competition between
manufacturers may pressure some to lower markups and absorb higher input costs. Thus, DOE
maintained the practice employed in previous rulemakings of applying a lower markup to the
incremental costs associated with more efficient products.
2.7

ENERGY USE ANALYSIS

The energy use analysis provides estimates of annual energy use for BCs and EPSs. To
estimate EPS energy use DOE multiplied annual usage in each application state and operating
mode (in hours per year) by the power estimates (in watts). For BCs, DOE multiplied annual
usage in each operating mode (in hours per year or charges per year) by the power estimates (in
watts) or energy per charge (in watt hours). DOE developed its power and energy estimates in

2-63

the engineering analysis. To determine annual usage, DOE examined how end users operate BCs
and EPSs with the consumer products they power.
DOE made the following key assumptions in calculating BC and EPS energy
consumption:
Energy consumption of an EPS or BC is defined as energy drawn from the source (e.g.,
mains) but not provided to the application (if an EPS) or battery (if a BC).
Usage and power requirements of the BC or EPS depend on usage and power
requirements of the application.
The power requirements of applications (and, in the case of BCs, batteries) do not change
over the period of analysis.
Only one usage profile is assigned to each application, although DOE considered
multiple user types in developing this profile for certain common applications.
While the mix of applications may change over time, the usage profile for the BCs or
EPSs in a product class as a whole remains fixed over time.
These assumptions are explained below and in chapter 7.
2.7.1 External Power Supply Usage and Energy Consumption
For most electrical appliances, power consumption is determined by measuring the power
an application draws from mains while performing its intended function(s). However, BCs and
EPSs are power conversion devices, and their intended function is to deliver a portion of the
power drawn from mains in one form to another application in another form. Therefore, the
traditional method for calculating energy consumption is not appropriate for BCs and EPSs.
Instead, energy consumption is determined by calculating the energy dissipated by the BC or
EPS.
An EPSs active mode power requirements depend on how much power the application
requires. Because the application has different operating states, the amount of power required
from the EPS varies by state. In order to examine EPS power consumption in active mode, DOE
determined the different operating states of the application and how much power (expressed as a
percentage of nameplate output power) each application state requires from the EPS.
During the framework document public meeting, DOE requested comment on its
assumption that most applications require 80 percent of the EPSs nameplate output power when
in the highest energy consuming application state. PTI commented that this assumption is a
reasonable estimate. (PTI, Pub. Mtg. Tr., No. 14 at p. 273) Subsequently, DOE tested four of the
most common EPS applications and determined that a lower default value may be more accurate.
DOE uses in its preliminary analysis a value of 60 percent of nameplate output power for those
applications for which data are lacking. DOE plans to test additional applications in the future
and may adjust this assumed value as appropriate.
EPS active mode power can easily be estimated for application states that demand 25
percent of nameplate output power or greater by using the average active-mode efficiency from
the test procedure to calculate the difference between the power drawn from mains and the

2-64

power provided to the application. However, EPSs are not as efficient when providing less than
25 percent of nameplate output power. Thus, DOE interpolated between EPS power at 25 percent
of nameplate output power and EPS power in no-load mode to more accurately estimate EPS
power at low loads. PG&E voiced support for this methodology, with PG&E noting that
Interpolation of a line between no load and 25% may not be accurate for all switch mode power
supplies. PG&E added that it believed that interpolation of the input power through this load
range is probably satisfactory. (PG&E, Pub. Mtg. Tr., No. 14 at p. 272; PG&E, No. 20 at p. 10)
DOE combined time and power estimates in active mode with time and power estimates
in no-load (standby) mode to calculate unit energy consumption in all EPS modes. Only EPSs
with on/off switches can operate in off mode. Because DOE did not identify any EPSs with
on/off switches, DOE did not attribute any time to this mode.
PTI commented in response to the NOPD that it agreed with the EPS usage profile DOE
used in its determination analysis for wall adapters involved in charging detachable batteries for
power tools. (PTI, EERE-2009-BT-DET-0005 No. 5 at p. 2) DOE carried forward that same EPS
usage profile into the preliminary analysis.
2.7.2 Battery Charger Usage and Energy Consumption
During the framework document public meeting, DOE suggested that it would create
usage profiles based on 24-hour charge cycles and use the 24-hour energy consumption
measurement from the test procedure to represent active mode energy consumption. DOE
received support from PG&E for basing usage profiles on 24-hour charge cycles. (PG&E, No. 20
at p. 9; PG&E, Pub. Mtg. Tr., No. 14 at p. 257) However, DOE determined that building usage
profiles around 24-hour charge cycles is unrealistic for frequently used applications such as MP3
players that are not left to charge for an entire day. Therefore, based on a suggestion by PG&E,
DOE isolated charge energy over the 24-hour charge cycle and examined BC usage in terms of
the number of complete charge equivalents and time spent in each mode annually. (PG&E, Pub.
Mtg. Tr., No. 14 at p. 257) DOE combined charges and charge energy with time and power
estimates for maintenance, no-battery (standby), and off modes to calculate unit energy
consumption in all BC modes.
DOE received input from three parties on BC usage profiles. PG&E provided usage
profiles for fourteen BC applications during the framework document comment period and later
submitted usage profiles with charge frequency for additional applications. (PG&E, No. 20 at p.
13) Philips described how usage profiles for personal care products differ, which allowed DOE
to develop separate usage profiles for beard trimmers and electric toothbrushes. (Philips, No. 22
at p. 3) PTI commented that BC usage profiles should account for multiple user types. (PTI, Pub.
Mtg. Tr., No. 14 at p. 259) Where data were available, DOE examined usage profiles for
different user types and developed a single usage profile to best represent overall usage for each
BC application.
2.8

LIFE-CYCLE COST AND PAYBACK PERIOD ANALYSIS

The LCC and PBP analysis is used to determine the economic impact of potential
standards on end users. The effects of standards on individual end users include changes in
2-65

operating expenses (usually lower) and changes in total installed cost (usually higher). DOE
analyzed the net effect of these changes for battery chargers (BC) and external power supplies
(EPS). DOE first calculated the changes in end-users LCCs likely to result from various
candidate standard levels compared to a base case (no new standards). The LCC calculation
considers total installed cost (which includes manufacturer selling price, sales taxes, distribution
chain markups, and installation cost), operating expenses (energy, repair, and maintenance
costs), product lifetime, and discount rate. DOE performed the LCC analysis from the
perspective of users of BCs and EPSs.
DOE also analyzed the effect of changes in operating expenses and installed costs by
calculating the PBP of potential standards relative to a base case. The PBP estimates the amount
of time it would take the end user to recover the higher purchase expense of more energy
efficient products through lower operating costs. Similar to the LCC, the PBP is based on the
total installed cost and operating expenses. However, unlike the LCC, only the first year's
operating expenses are considered in the PBP calculation. Because the PBP does not account for
changes in operating expense over time or the time value of money, it is also referred to as a
simple payback period. For more detail on the LCC and PBP analyses, see chapter 8 of the
preliminary TSD.
2.8.1 Approach
DOE conducted the LCC and PBP analysis by modeling both the uncertainty and
variability in the inputs using Monte Carlo simulation methods and probability distributions.
(See appendix 8B of the preliminary TSD for a detailed explanation of Monte Carlo simulation
and the use of probability distributions). DOE implemented Monte Carlo simulation and
probability distributions by using Microsoft Excel spreadsheets combined with Crystal Ball, a
commercially available add-in program.
DOE defined the use characteristics of BCs and EPSs by characterizing a set of end-use
applications (applications) for each representative unit and representative product class
(representative unit). These applications determine the profiles of the BCs and EPSs, such as
energy usage, product lifetime, and pricing markups.
DOE used data from the Energy Information Administration (EIA) on electricity prices
for different customer classes and electrical utilities to establish the variability in energy pricing.
Due to the large range of applications and product use characteristics considered in the LCC and
PBP analysis, the range of annual energy use is broad. Although the annual energy use and/or
energy pricing are known for any particular application, the variability across all applications
contributes to the range of LCCs and PBPs calculated for any particular standard level.
In each Monte Carlo simulation, a typical energy use profile is selected from a
distribution of energy consumption profiles for the different applications associated with each
representative unit. The selected profile for the representative unit is used to estimate the BC or
EPS energy use. Further information on these application profiles can be found in appendix 8C
of this preliminary analysis.

2-66

2.8.2 Life-Cycle Cost Inputs


For each efficiency level analyzed, the LCC analysis requires input data for the total
installed cost of the BC or EPS, the operating cost, and the discount rate. Table 2.40 summarizes
the inputs DOE used to calculate the total installed cost and operating cost of various energy
efficiency levels for BCs and EPSs. A more detailed discussion of the inputs follows. For more
information on data sources and assumptions used, see chapter 8 of the preliminary TSD.
Table 2.40 Summary of Inputs Used in the Life-Cycle Cost Analysis
Input

Characterization

Manufacturer Selling Price

Varies with BC and EPS

Markups

Varies with BC and EPS

Sales Tax

Varies by census region

Installation Cost

No installation cost associated with BCs or EPSs

Unit Energy Consumption

Varies with BC and EPS by efficiency level

Energy Prices

Vary by sector and census region

Energy Price Trends

Vary with price forecast scenario

Repair and Maintenance Costs No repair or maintenance costs associated with BCs or EPSs
Lifetime

Mean lifetime for a BC or EPS varies with application

Discount Rate

Varies with sector

2.8.2.1 End-User Product Price


BCs and EPSs are unique appliances because they are always used in conjunction with
other products of interest. Since most BCs and EPSs are packaged with particular products,
consumers usually do not buy the BC or EPS directly. Instead, consumers typically obtain an
EPS for a video game system when buying the video game system itself. Although the LCC and
PBP analysis uses the consumer purchase prices of BCs and EPSs, in reality those prices are
most often a hidden portion of the prices that consumers pay for the products of interest.
DOE determined the baseline and incremental markups for each point in the BC and EPS
supply chain, as well as the sales tax, in the markups analysis (chapter 6 of the preliminary TSD).
The types and amounts of markups differ between applications categorized by the representative
units. For this reason, DOE calculated markups independently for each application considered. A
listing of total markups by application as used in the LCC and PBP analyses can be found in
appendix 8C to this preliminary analysis.
2.8.2.2 Distribution Markups and Sales Taxes
DOE determined the baseline and incremental markups for each point in the BC and EPS
supply chain, as well as the sales tax, in the markups analysis (chapter 6 of the preliminary TSD).
The types and amounts of markups differ between applications categorized by the representative
units. For this reason, DOE calculated markups independently for each application considered. A
listing of total markups by application as used in the LCC and PBP analyses can be found in
appendix 8C to this preliminary analysis.

2-67

2.8.2.3 Installation Costs


DOE found no relevant installation costs associated with BCs or EPSs. This is because
installation typically entails a consumer simply unpacking the BC or EPS from the box it was
sold in and connecting the device to mains power and its associated product or battery. Because
the cost of this installation (which may be considered temporary, as intermittently used devices
might be unplugged for storage) is not quantifiable in dollar terms, DOE considers the
installation cost to be zero. For this reason, DOE categorizes the total installed cost of BCs and
EPSs as the final product price for these units.
2.8.2.4 Energy Prices and Energy Price Trend
DOE developed nationally representative distributions of electricity prices for different
customer categories (residential and commercial) from 2008 EIA Form 861 data. At the
framework document public meeting, Vivitar commented that a recent news article noted that
many users charge their products while at work, and suggested that DOE consider sectors other
than the residential sector for its electricity price analyses. (Vivitar, Pub. Mtg. Tr., No. 14 at p.
349) DOE agrees with this comment, and has expanded its sector analysis to include the
commercial sector for the life-cycle cost analyses. At this time, DOE does not have sufficient
data to analyze how residential consumers use their BC or EPS in a commercial setting, or vice
versa. The average electricity prices for each sector (in 2009$) are 11.5 cents for the residential
sector, and 10.1 cents for the commercial sector.
PG&E commented that DOE should use marginal electricity prices rather than average
prices because they most closely represent the actual changes on a consumers energy bill.
(PG&E, Pub. Mtg. Tr., No. 14 at pp. 358, 356, 374; PG&E, No. 20 at p. 10) DOE notes that
average electricity prices are typically used for rulemakings on products that do not contain a
great degree of seasonality or time sensitivity, whereas marginal electricity prices are typically
used for rulemakings on products that do. This is because the marginal energy prices can help
capture insight into energy prices during peak loading hours for products that are used differently
at different periods of time. DOE has found no evidence that this is the case for BCs and EPSs,
which operate without much seasonality or time sensitivity. In cases such as these, the energy
costs from marginal electricity rates are oftentimes close to that of average prices. In light of this
fact and the complexity of incorporating marginal electricity rates, DOE decided to use average
electricity prices for its BC and EPS analyses. However, DOE may consider a consumer
subgroup with steeply inclined block rates during the NOPR stage of the rulemaking, as
described in section 2.11.
Similarly, PG&E commented that DOE should factor time of use into its electricity prices
because California is moving toward critical-peak pricing, and other parts of the country are
looking into a time-of-use system as well. (PG&E, Pub. Mtg. Tr., No. 14 at p. 360) DOE has not
found any data to suggest that time-of-use factors have a significant impact on BCs or EPSs.
Additionally, DOE notes that only a portion of the nations utilities charge customers based on a
time-of-use plan. Due to the fragmented nature of the utilities industry, it would be difficult to
draw conclusions based on the limited time-of-use data available. As a result, DOE did not
consider time-of-use when calculating electricity prices.

2-68

During the public meeting, PG&E inquired if DOE planned to use energy prices from a
nationally weighted average by population or a simple average. (PG&E, Pub. Mtg. Tr., No. 14 at
p. 357) Consistent with other rulemakings, DOE utilized a national utility-consumer weighted
average for determining energy prices. Additionally, DOE examined consumer-weighted
electricity prices broken out at different regional levels for its Monte Carlo simulation analysis.
DOE used recent price forecasts by EIA to estimate future trends in electricity prices in
each sector. To arrive at prices in future years, DOE multiplied the average prices described in
the preceding section by the forecast of annual average price changes in EIAs Annual Energy
Outlook (AEO) 2010 Early Release. To estimate the trend after 2035, DOE followed past
guidelines provided to the Federal Energy Management Program (FEMP) by EIA and used the
average rate of change from 2025 to 2035 for electricity.
During the public meeting, PG&E asked if DOE planned to conduct interviews with
utilities to try to get information about future rate increases. (PG&E, Pub. Mtg. Tr., No. 14 at p.
361) PG&E also commented that DOE should account for Renewable Energy Portfolio
Standards (RPS) when calculating future energy prices. (PG&E, Pub. Mtg. Tr., No. 14 at p. 373)
DOE decided to utilize the AEO for projecting energy price trends, as is the standard procedure,
and the AEO already takes these two suggestions into account. The projections in the AEO
consider a comprehensive review of utilities, and also consider the ramifications of the statelevel RPSs.
DOE calculated the LCC and PBP using three separate projections from AEO 2010:
Reference, Low Economic Growth, and High Economic Growth. The AEO2010 Early Release
was the most recent AEO available at the time of this preliminary analysis. This AEO2010
contained only the reference-case economic growth projection, however. In order to develop
high-economic-growth and low-economic-growth cases for the LCC and PBP analysis, DOE
obtained the differences between the reference case and the high- and low-economic-growth
cases in the March 2009 AEO2009, which includes all three cases. DOE then utilized these
differences as scaling factors to generate high- and low-economic-growth cases based on the
AEO2010 Early Release reference case.
At the public meeting, PG&E commented that DOE should consider environmental
impacts, such as cap and trade emissions limitations. (PG&E, Pub. Mtg. Tr., No. 14 at p. 358) In
response to this comment, DOE also considered an electricity price projection sensitivity using a
carbon cap and trade scenario developed in response to the Lieberman-Warner Climate Security
Act of 2007. This sensitivity illustrates an elevated electricity price projection based on
emissions regulations as outlined in the EIAs S.2191 report accompanying the Act.
2.8.2.5 Repair and Maintenance Costs
BCs and EPSs are not usually repaired because they typically outlast the applications that
they power, and the cost of repair would generally exceed the cost of a new replacement. DOE
found no evidence that repair or maintenance costs would increase with higher BC or EPS
energy efficiency. Thus, DOE did not include changes in repair and maintenance costs for BCs
or EPSs that are more efficient than baseline products.

2-69

2.8.2.6 Lifetime
DOE considers the lifetime of a BC or EPS to be from the moment it is purchased for end
use up until the time when it is permanently retired from service. Because the typical BC or EPS
is purchased for use with a single associated application, DOE assumes that it will remain in
service for as long as the application does. Therefore, the lifetime of a given BC or EPS is
directly tied to the useful life of the application it is associated with, and will not vary with
increases in energy efficiency. In instances where the lifetime of the BC or EPS is not tied to the
application, DOE will apply the lifetime of the BC or EPS rather than the application lifetime. A
possible example may be the lifetime of a cell phone charger, as manufacturers move towards
implementing a universal charger for cell phones.
Each representative unit for BCs and EPSs contains several applications, and each
application has its own, unique lifetime. In light of this, DOE sampled lifetimes from among the
applications associated with each representative unit using their shipments for weighting.
Accordingly, the lifetime distribution for each representative unit is based on the importance of
each application to the representative unit, as determined by shipments. The lifetimes for these
applications were taken from market data and industry interviews for the relevant applications,
and are outlined in chapter 3 of the preliminary analysis. DOEs application sampling
methodology is covered in more detail in appendix 8C of this preliminary analysis. The
summarized lifetime estimates for each representative unit are outlined in Table 2.41.
Table 2.41 Market-Weighted Lifetimes for Representative Units
Representative Unit

Lifetime
Yrs

Class A External Power Supplies


2.5 Watt Regular AC/DC

4.6

18 Watt Regular AC/DC

4.4

60 Watt Regular AC/DC

4.0

120 Watt Regular AC/DC

3.3

Battery Chargers
Low Energy, Inductive

5.0

Low Energy, Low Voltage

3.4

Low Energy, High Voltage

4.2

Medium Energy, Low Voltage

5.5

High Energy

6.4

Low Energy, AC Out

6.0

2.8.2.7 Discount Rate


The discount rate is the rate at which future expenditures are discounted to estimate their
present value. DOE derived the discount rates separately for residential and commercial
consumers. For residential consumers, DOE estimated the discount rate by looking across all
possible debt or asset classes that might be used to purchase or operate BCs or EPSs. For the
commercial consumers, DOE estimated the cost of capital for commercial companies by

2-70

examining both debt and equity capital and developing an appropriately weighted average of the
cost to the company of equity and debt financing. For each end-use sector, DOE developed a
distribution of discount rates from which the Monte Carlo simulations sample. Further detail on
the discount rate calculations can be found in chapter 8 of the preliminary TSD.
For the residential sector, DOE assembled a distribution of interest or return rates on all
possible debt or asset classes that might be used to purchase BCs or EPSs. DOE estimated the
average shares of the various debt and asset classes in the average U.S. household portfolio using
the Federal Reserve Boards Survey of Consumer Finances (SCF) data for 1989, 1992, 1995,
1998, 2001, 2004, and 2007. DOE used the mean share of each class across the seven survey
years (18 years) as the basis for estimating household financing of BCs and EPSs. The weightedaverage discount rate for residential product owners is 5.6 percent. Further detail on the
residential discount rate calculations can be found in appendix 8D of the preliminary TSD.
For the commercial sector, DOE used a weighted average of the cost to the firm of equity
and debt financing. DOE estimated the cost of equity financing using the Capital Asset Pricing
Model (CAPM). The CAPM, among the most widely used models to estimate the cost of equity
financing, assumes that the cost of equity is proportional to the amount of systematic risk
associated with a firm. DOE estimated the cost of debt financing by adding a risk adjustment
factor to the current yield on long-term corporate bonds (the risk-free rate). This procedure is
used to estimate current and future company costs to obtain debt financing.
By calculating interest rates for equity and debt financing, DOE was able to determine a
distribution of discount rates for each class of potential owners using data from the Damodaran
online investment survey.i Each class of potential owners was then weighted by the number of
total employees in that class, using comparative statistics from the U.S. Census Bureau. The
weighted-average discount rate for commercial product owners is 7.0 percent.
At the July 16, 2009, public meeting, PG&E asked if the low interest rates present in the
current economy would artificially skew the discount rate for the consumer, and how DOE
accounts for the value of money to a consumer in a recession. (PG&E, Pub. Mtg. Tr., No. 14 at
p. 354) DOE does not believe that its discount rates are artificially skewed based on the current
economic conditions because the rates are based on data spanning 20 to 30 years. Consistent with
other rulemakings, DOE calculates the discount rate by analyzing multiple forms of debt and
equity financing from 1989 through 2007 (the most current data), in addition to the opportunity
cost of multiple asset classes and their returns over the time period of 1979 to 2009. This method
of calculating a discount rate ensures that temporary disruptions in interest rates will not have a
significant impact on the discount rate used during the analysis.
2.8.2.8 Compliance Date of Standard
The compliance date is the future date when a new standard becomes operative. DOEs
publication of a final rule in this standards rulemaking is scheduled for completion by July 2011.
The compliance date for amended EPS standards is July 1, 2013. (42 U.S.C.
6295(u)(3)(D)(i)(II)(bb)) The compliance date for BC standards is also July 1, 2013. DOE
i

The survey is available at http://pages.stern.nyu.edu/~adamodar.

2-71

calculated the LCCs for all consumers as if each would purchase a new product in the year of the
compliance date (January 1, 2013). However, DOE based the cost of the product on the most
recent available data. All dollar values are expressed in 2009 dollars.
2.8.2.9 Product Energy Efficiency in the Base Case
For the purpose of conducting the LCC analysis, DOE analyzed candidate standard levels
relative to a base case (i.e., a case without new energy conservation standards). This required
estimating the distribution of product efficiencies in the base case (i.e., what consumers would
have purchased in 2013 in the absence of new standards). Rather than assuming that all
consumers will purchase products at the baseline efficiency level when it analyzed the impacts of
a particular standard level, DOE conducted the analysis to account for the wide range of product
energy efficiencies that consumers are expected to purchase under the base case.
DOEs approach for conducting the LCC analysis for BCs and EPSs relied on developing
samples of consumers that use each of the products and using a Monte Carlo simulation
technique to perform the LCC calculations on the consumers in the sample. DOE assigned each
consumer in the sample a unique product energy efficiency level taken from the estimated base
case distribution of product energy efficiencies in the compliance year. The energy efficiency
distributions used for each application and each representative unit are presented in chapter 3 of
the TSD. The applications assigned to each representative unit and their respective energy
efficiency distribution are explained in further detail in Appendix 8C.
DOE calculated the efficiency distributions by evaluating tested efficiency data for each
representative unit. The tested efficiencies were then binned into the appropriate standards level
and weighted by the shipments of the application tested. DOE assumed that the current averages
for BCs are reasonable to use as a base case for 2013, but assumed a modest improvement in
EPS efficiency from 2009 to 2013. This is because efficiency standards requiring EPSs to meet
an ENERGY STAR level (CSL 1) will take effect in year 2010 in Europe. DOE assumes this
will impact the American market by shifting EPS efficiencies higher by year 2013. Further detail
on this analysis can be found in chapter 9 of the TSD.
2.8.3 Payback Period
As described above, the PBP is the amount of time it takes the customer to recover the
assumed additional installed cost of more efficient products through energy cost savings.
Payback analysis is a common technique used to evaluate investment decisions. This calculation
is known as a simple payback period because it does not account for changes in operating
expense over time or the time value of money.
The inputs to the PBP calculation are the total installed costs of the product to the end
user for each efficiency level and the annual (first-year) operating expenditures for each
efficiency level. The PBP calculation uses the same inputs as the LCC analysis, except that
energy price trends and discount rates are not needed. The calculation uses energy prices only for
the year in which a new standard is expected to take effect, in this case 2013.

2-72

2.8.4 Sensitivity Analysis


In addition to the Monte Carlo simulations used to characterize uncertainty and variation
among the input values, DOE also considered two alternate sensitivity scenarios in the LCC and
PBP analysis. The two sensitivities considered a positive or negative shift in the mean value
associated with several key inputs, including: MSPs, lifetimes, usage profiles, and loading points
(for EPSs only). DOE believes these sensitivities characterize the diversity of available data, and
presents the results to highlight the impacts this diversity can have on the LCC and PBP.
For the two sensitivity cases shown in Table 2.42, DOE simultaneously varied the inputs
to obtain the highest and lowest possible LCC savings that could be reasonably expected based
on the diversity seen in the data. While the reference case LCC results (explained previously)
examine variability around a mean value for several inputs, the high-savings and low-savings
sensitivities consider variability around a reasonably higher or lower mean input value.
Appendix 8B presents the input values and resulting LCC and PBP projections for the highsavings and low-savings sensitivities.
Table 2.42 Input Variations for the Sensitivity Analysis
Input
MSPs

Reference Case
Original values

Lifetimes
Usage Profiles

Original values
Original values

Low Savings Case


Higher costs
(MSP + average
percentage error)
Shorter lifetimes
Less time in use

Loading Points
(for EPSs only)

Original values
(60 percent for highest
active mode state where
no test data available)

Lower values
(40 percent for highest
active mode state where
no test data available)

2.9

High Savings Case


Lower costs
(MSP average
percentage error)
Longer lifetimes
More time in use
Higher values
(80 percent from nonClass A EPS
determination for highest
active mode state where
no test data available)

SHIPMENTS ANALYSIS

The shipments analysis generates shipment forecasts and efficiency forecasts for each of
the BC and EPS product classes being considered for standards. The shipment forecasts are used
in the NIA (chapter 10) and the MIA (chapter 12). The efficiency forecasts are used in the LCC
(chapter 8), and the NIA.
DOE made the following key assumptions in conducting the shipments analysis:

The market for BCs and EPSs will grow at the same rate as the resident population

over the next 33 years.

Half of the Class A EPS market at CSL 0 in 2009 will have transitioned to CSL 1 by

2013.

2-73

BCs will not become more efficient between 2009 and 2013.
Products would not become more efficient after 2013 in the absence of new or

amended Federal standards.

Standards would have no impact on BC and EPS shipment volumes.

Manufacturers would respond to standards by improving those products that do not

meet the standards, while the products that were already as or more efficient than the

standard would remain unaffected.

These assumptions are explained below and in chapter 9.


2.9.1 Shipments Forecast
The shipments forecast gives the total number of BCs and EPSs shipped each year. To
create this forecast, DOE combined current year shipments, discussed in the market assessment
(chapter 3), with a compound annual growth rate for BCs and EPSs and generated unit shipment
values through the analysis period.
2.9.1.1 Compound Annual Growth Rate
DOE faces several challenges in calculating a compound annual growth rate. First, DOE
must conduct its shipments forecast to account for 30 years of shipments. In addition, DOE must
consider the entire potential market for BCs and EPSs, which could both include products that do
not yet exist and exclude current products that may be discontinued before or during the analysis
period. In addition, DOE found that many existing forecasts project out for only a few years, and
most cover only a small number of BC and EPS applications.
In light of these challenges, DOE concluded that the best methodology for approximating
future market growth in the preliminary analysis would be to hold constant the current (2008)
application mix and per-capita purchase rates for BCs and EPSs and account only for population
growth. These assumptions are explained further in chapter 9.
2.9.1.2 Impact of Standards on Shipments Forecast
DOE considered the impact of standards on shipments in the MIA (chapter 12). In that
analysis, manufacturers predicted that standards would have no negative impact on consumer
consumption of EPSs due to cost increases. In addition, DOE has determined that the utility and
functionality of an EPS minimizes the potential for substitution of alternative power sources.
Therefore, DOE assumed that standards would have no impact on the volume of EPS shipments.
DOE extended this assumption to hold true for BCs as well.
2.9.2 Efficiency Forecast
The efficiency forecast gives the distribution of EPS and BC shipments by CSL in the
absence of standards (the base case). To develop its efficiency forecast, DOE first assessed
present-day (2009) efficiency and then considered how the efficiency of new units might change
by the first year of the analysis period (2013) and throughout the analysis period in the absence
of new or amended Federal standards.

2-74

2.9.2.1 Base Year Efficiency Distribution


DOE estimated the efficiency distribution in 2009 for each product class using currently
available market and test data, as discussed in chapter 3.
2.9.2.2 Effects from Outside EPS and BC Standards
DOE considered the impact of outside standards for BCs and EPSs on the efficiency
distribution in 2013 and beyond.
For BCs, DOE found that, with the exception of Canada, there were no BC standards
slated to take effect by 2013, and minimal participation by BC manufacturers in voluntary
programs such as ENERGY STAR. In light of this, DOE could find no compelling evidence that
battery charging systems will improve in efficiency in the short term and, consistent with these
facts, assumed that the current efficiency distributions for BCs would be reasonable to use as a
base case forecast for the entire analysis period.
For EPSs, however, DOE found some programs, both voluntary and mandatory, that are
likely to influence EPS efficiency in the short term. These programs, such as ENERGY STAR in
the United States (voluntary) and the ecodesign requirements for external power supplies (EC
No. 278/20093) in Europe (mandatory), specify that Class A EPSs meet or beat CSL 1. With this
in mind, and given the global nature of the EPS manufacturing industry, DOE estimated that in
the United States approximately half of the Class A EPS market at CSL 0 in 2009 would
transition to CSL 1 by 2013. DOE will update and refine this estimate in advance of publishing
the NOPR.
2.9.2.3 Effects from Current Class A EPS Standards
In the framework document, DOE presented its preliminary assessment of the potential
for current Federal standards for Class A EPSs to affect non-Class A EPS efficiency. DOE noted
that the construction of certain non-Class A EPSs (specifically EPSs for medical devices and
certain battery-charged products) was similar to that of Class A EPSs and that these similarities
made it likely that efficiency improvements for Class A EPSs would transfer to these other EPSs.
Therefore, in the present analysis DOE assumes that by the first year of the analysis period
(2013) EPSs for medical devices and certain battery charged products will be just as efficient as
Class A EPSs were in 2009.
During the framework document public meeting, Micro Power Electronics, which
primarily serves the commercial, medical, and military markets, commented that EPS efficiency
standards that started in the household-use area (like ENERGY STAR and CEC) have migrated
into our market areas. (MPE, No. 5 at p. 1) DOE believes that this comment provides support
for its assumption that the standards currently in effect for Class A EPSs will lead to efficiency
improvements in certain non-Class A EPSs as well.
2.9.2.4 Market Response to Standards
DOE considered how the mix of product efficiencies might change due to new or
amended Federal standards. In its preliminary analysis, DOE assumed that manufacturers would
2-75

respond to standards by improving those products that do not meet the standards to meet the
standard level, but not higher, while the products that were already as or more efficient than the
standard would remain unaffected.
2.10

NATIONAL IMPACT ANALYSIS

The national impact analysis estimates energy savings and assesses the NPV of consumer
LCC savings at the national scale. The results can be used to identify the CSL that, for a given
product class, yields the greatest energy savings while remaining cost effective from a consumer
perspective. DOE estimated both NES and NPV for all candidate standard levels for each BC
and EPS product class. To make the analysis more accessible and transparent to all interested
parties, it is documented in a Microsoft Excel spreadsheet model that can be downloaded from
the EERE website.
The NIA considers total installed cost (which includes manufacturer selling prices,
distribution chain markups, sales taxes, and installation costs), operating expenses (energy,
repair, and maintenance costs), product lifetime, and discount rate. However, where the LCC
considers the savings and costs associated with standards for a set of representative units, the
NIA considers the savings and costs associated with all units affected by standards during the
entire analysis period. Chapter 10 provides additional details regarding the NIA.
DOE made the following key assumptions in conducting the NIA:
Products remain in the installed base for an integral number of years (no fractional

years)

All energy cost savings accrue at residential electricity prices.

These assumptions are explained below and in chapter 10.

DOEs estimates of NES and NPV for BCs and EPSs should not be added to one another
because many BCs incorporate an EPS. One way to improve the efficiency of these BCs is to
improve the efficiency of the EPS part. The resulting savings are counted in both the BC analysis
and in the separate EPS analysis. Thus, due to this overlap, combining BC and EPS NES (or
NPV) estimates would overstate savings resulting from improving BCs and EPSs.
2.10.1 Approach
DOE conducted the NIA by combining unit costs and unit energy consumption with
efficiency distributions and shipments over a shipment forecast period of 30 years and an
analysis period extending until all units shipped during the shipment forecast period are retired
from use. Energy savings were compared between a base case and multiple standards cases,
yielding NES estimates. Unit improvement costs and energy cost savings were then discounted
to the present day (2010) and compared to one another, yielding NPV estimates.

2-76

DOE conducted an NIA sensitivity analysis that mirrored the LCC sensitivity analysis,
which is explained briefly in section 2.8.4. For additional explanation of how inputs varied
between the low-savings, reference, and high-savings scenarios see appendix 8B.
2.10.2 NIA Inputs
The NIA requires a number of inputs, summarized in Table 2.43. Many are the same as
those required for the LCC. These shared inputs are detailed in Section 2.8.2. Variations from the
LCC include the use of different discount rates and the extension of shipments and efficiency
forecasts beyond 2013.
Table 2.43 Summary of Inputs Used in the National Impact Analysis
Input

Characterization

Manufacturers Selling Price

Varies with BC and EPS product class, from chapter 5.

Markups

Varies with BC and EPS product class, from chapter 6.

Sales Tax

U.S. National Average, from chapter 6.

Installation Cost

No installation cost associated with BCs or EPSs.

Unit Energy Consumption

Varies with BC and EPS product class and by efficiency level, from chapter 7.

Electricity Prices and Trends

U.S. National Average Residential, varied with forecast scenario, from chapter 8.

Site-to-Source Conversion

U.S. National Average Rate.

Repair and Maintenance Costs No repair or maintenance costs associated with BCs or EPSs.
Shipments

Varies with BC and EPS product class, from chapter 9.

Lifetime

Varies with BC and EPS product class, from chapter 3.

Discount Rate

3% and 7%, as specified by the Office of Management and Budget.

Efficiency Distribution in
2013

Varies with BC and EPS product class, from chapter 9.

As noted above, while the LCC analysis weighs the costs and savings associated with
standards for a set of representative units, the NIA considers the costs and savings associated
with standards for all products, including those in the scaled product classes. For these scaled
product classes, most inputs to the NIA could be obtained in the same manner as for the
representative product classes. However, cost data were not available for the scaled product
classes, as they were not directly analyzed in the engineering analysis. Therefore, as is done in
other rulemakings, DOE extrapolated from its knowledge of costs for the representative product
classes to the scaled product classes. Costs from the 2.5 W EPS representative unit were applied
to product classes A2 and A4, and costs from the 18 W EPS representative unit were applied to
product class A3. Costs for BC product classes 3, 6, 8, and 9 were derived from cost data for the
representative product classes as described in chapter 5.
2.10.3 National Energy Savings Analysis
The major inputs for determining the NES for each product analyzed are annual unit
energy consumption, shipments, lifetimes, and site-to-source conversion factors. DOE calculated
national energy consumption for each year by multiplying unit energy consumption by the
2-77

number of units in the installed base in that year. NES for a given year, then, is the difference in
national energy consumption between the base case (without new efficiency standards) and each
standards case. DOE estimated energy consumption and savings first in terms of site energy and
then converted the savings into source energy. Cumulative energy savings are the sum of the
NES estimates for each year.
2.10.4 Net Present Value Analysis
The major inputs for determining NPV are installed costs, annual operating cost savings,
and discount rates. For each standards case, DOE calculated net savings each year as the
operating cost savings less any increase in installed costs due to the standard. DOE calculated
operating cost savings over the life of each product, accounting for changes in electricity prices
over time. DOE calculated NPV as the present value of operating cost savings less the present
value of increased installed costs. DOE used discount rates of 3 percent and 7 percent to discount
future costs and savings to present values to reflect the time value of money.
In the preliminary analysis, DOE assumed that all energy cost savings will accrue at
residential electricity prices. Since commercial electricity prices are lower, the national energy
cost savings estimates are slightly overestimated to the extent that some BCs and EPSs are used
in the commercial sector. The NIA workbook contains both residential and commercial price
forecasts and a control for choosing a sector split, should sensitivity analysis be necessary in the
NOPR stage.
2.11

LIFE-CYCLE COST SUBGROUP ANALYSIS

The LCC subgroup analysis will evaluate consumer impacts by identifying consumer
subgroups and accounting for variations in key inputs to the LCC analysis. A consumer subgroup
comprises a subset of the population that is likely, for one reason or another, to be impacted
disproportionately by new or revised energy conservation standards. The purpose of a subgroup
analysis is to determine the extent of this disproportional impact. During the NOPR stage, DOE
intends to evaluate the impacts of particular subgroups of customers in part by analyzing the
LCC for these particular customers.
DOE intends to evaluate variations in usage profiles and regional electricity prices, such
as customers with steeply inclined marginal block rates (resulting in electricity prices higher than
the average price), to see what type of effect an energy efficiency standard would have on the
NPV for these customer subgroups. Additionally, DOE may consider evaluating LCC impacts on
a subgroup of applications within a representative unit. To the extent possible, DOE plans to
determine estimates of the variability of each input parameter and consider this variability in its
calculation of customer impacts. DOE intends to perform sensitivity analyses to consider how
differences in energy use will affect various subgroups of customers.
DOE will determine the effect on customer subgroups using the LCC spreadsheet model.
The spreadsheet model used for the LCC analysis can be used with different data inputs. The
standard LCC analysis includes various customer types that use BCs and EPSs. DOE can analyze
the LCC for any subgroup by using the LCC spreadsheet model and sampling only that
subgroup. The model is explained in chapter 8 of the preliminary TSD, which describes the LCC
2-78

and PBP analyses. DOE will be especially sensitive to potential negative economic impacts on
identifiable population groups such as small businesses (i.e., those with low annual revenues),
which may not be able to afford a substantial increase in the price of BCs or EPSs.
PG&E commented that DOE should especially consider the effect of standards on low
income users and small businesses. (PG&E, Pub. Mtg. Tr., No. 14 at pp. 46, 351) PG&E also
commented that DOE should consider consumers with electricity prices that are higher than the
average rate of electricity, which it suggests may be more likely for BC and EPS users. (PG&E,
Pub. Mtg. Tr., No. 14 at p. 363) One way that PG&E suggested doing this would be to perform
regional sensitivity using top-tier residential electricity prices rather than average electricity
rates. (PG&E, No. 20 at pp. 10, 358) DOE acknowledges these comments, and will revisit them
in the NOPR.
2.12

PRELIMINARY MANUFACTURER IMPACT ANALYSIS

The purpose of the MIA is to identify the likely impacts of higher energy conservation
standards on manufacturers. DOE will identify and discuss a wide range of quantitative and
qualitative industry impacts in interviews with manufacturers and other interested parties during
the NOPR stage of the analysis. DOE will use the manufacturer interviews as well as the
methodology described below in its evaluation of standards for BCs and EPSs.
DOE will conduct the MIA in three phases, and will further tailor the analytical
framework based on the comments it receives. In Phase I, DOE creates an industry profile to
characterize the industry and identify important issues that require consideration. In Phase II,
DOE prepares an industry cash-flow model and an interview questionnaire to guide subsequent
discussions. In Phase III, DOE interviews manufacturers and assesses the impacts of standards
both quantitatively and qualitatively. DOE assesses industry and subgroup cash flow and NPV
using the Government Regulatory Impact Model (GRIM). DOE then assesses impacts on
competition, manufacturing capacity, employment, and regulatory burden based on manufacturer
interview feedback and discussions.
Until recently, DOE reported MIA results in its standards rulemakings only in the NOPR
stage of the rulemaking. However, DOE is now evaluating and reporting preliminary MIA
information at this preliminary analytical stage. DOE gathered the information for the analysis
during the manufacturer interviews conducted after the engineering analysis. See chapter 12 of
the preliminary TSD for more detail on the MIA.
2.12.1 Sources of Information for the MIA
Many of the analyses described above provide important information applicable to the
MIA. Such information includes manufacturing costs and prices from the engineering analysis,
as well as retail price and shipment forecasts. DOE will supplement this information with
company financial data and other information gathered during interviews with manufacturers.
The interview process plays a key role in the MIA because it allows interested parties to
privately express their views on important issues. DOEs contractor has a non-disclosure
agreement in place with interviewees to allow consideration of confidential or sensitive
2-79

information in the decision-making process. Confidential information will not be made available
in the public record. To preserve confidentiality, DOEs contractor aggregates these perspectives
across manufacturers, creating a combined opinion or estimate for DOE to use in analyzing the
impacts standards will have on manufacturers. This process enables DOE to account for sensitive
information shared by manufacturers during the rulemaking process.
The process also includes detailed interviews with manufacturers to gain insight into the
range of potential impacts of standards. Typically, the interviews solicit both quantitative and
qualitative information on the potential impacts of efficiency levels on sales, direct employment,
capital assets, and industry competitiveness. An interactive interview process, rather than a
written response to a questionnaire, is used because it helps clarify responses and identify
additional issues. Before the interviews, a draft document showing estimates of the financial
parameters based on publicly available information will be circulated. Comments on these
estimates will also be solicited during the interviews. See chapter 12 of the preliminary TSD for
more detail on the methodology used in the MIA.
2.12.2 Industry Cash-Flow Analysis
The industry cash-flow analysis relies primarily on the GRIM. DOE uses the GRIM to
analyze the financial impacts of more stringent energy conservation standards on the industry.
The GRIM analysis uses several factors to determine annual cash flows from an amended
energy conservation standard: annual expected revenues; manufacturer costs, including cost of
goods sold, depreciation, research and development, selling, and general and administrative
expenses; taxes; and conversion capital expenditures. DOE compares the results against base
case projections that involve no amended energy conservation standards. The financial impact of
amended energy conservation standards is the difference between the two sets of discounted
annual cash flows. Other performance metrics, such as return on invested capital, also are
available from the GRIM. See chapter 12 of the preliminary TSD for more information on the
industry cash-flow analysis.
2.12.3 Manufacturer Subgroup Analysis
Industry cost estimates are inadequate to assess differential impacts among subgroups of
manufacturers because these subgroups may have different cost structures or regulatory
frameworks that affect their respective business models. For example, small and niche
manufacturers, or manufacturers whose cost structure differs significantly from the industry
average, could experience a more negative impact. While considering impacts on an individual
firm basis would be ideal, such an approach is typically impractical given the number of firms
involved. In place of this approach, DOE typically groups manufacturers exhibiting similar
characteristics together to characterize the industry.
During the interview process, DOE will discuss the potential subgroups and subgroup
members it has identified for the analysis. DOE will encourage manufacturers to recommend
subgroups or characteristics that are appropriate for the subgroup analysis. See chapter 12 of the
preliminary TSD for more detail on the manufacturer subgroup analysis.

2-80

2.12.4 Competitive Impacts Assessment


Section 325(o)(2)(B)(i)(V) of EPCA (codified at 42 U.S.C. 6295(o)(2)(B)(i)(V)) directs
DOE to consider any lessening of competition likely to result from imposition of standards. DOE
will consider whether an amended energy conservation standard is likely to reduce industry
competition and will ask the Attorney General to determine the impacts, if any. The assessment
will be based on manufacturing cost data and information collected from manufacturer
interviews. DOE will make a determined effort to gather and report firm-specific financial
information and impacts.
2.12.5 Cumulative Regulatory Burden
DOE recognizes and seeks to mitigate the overlapping effects on manufacturers of new or
amended DOE standards and other regulatory actions affecting the same equipment. DOE will
analyze and consider the impact on manufacturers of multiple, equipment-specific regulatory
actions.
During the public meeting, the American Council for an Energy-Efficient Economy
commented that there is current legislation in Congress intended to help manufacturers offset
conversion costs. DOE should take this legislation into account if it is passed. (ACEEE, Pub.
Mtg. Tr., No. 14 at p. 343)
In its response to these comments, DOE stated that although it continues to track
upcoming legislation, it cannot incorporate legislative proposals into its analysis until the
proposals become law.
2.12.6 Preliminary Results for the Manufacturer Impact Analysis
DOE conducted a series of preliminary interviews with manufacturers to discuss impacts
that may result from regulatory standards for battery chargers and external power supplies.
Related to standards for external power supplies, manufacturers and other parties identified four
general areas of concern: (a) impacts on shipments and (b) timing, (c) efficiency levels, and (d)
regulatory approval. These concerns are described briefly in the following paragraphs; please
see TSD chapter 12 for a more complete description of these concerns.
2.12.7 Impacts on Shipments
Manufacturers stated that if exceptionally stringent standards caused costs and prices to
rise, shipments could be affected.
2.12.8 Timing
Some manufacturers commented that if standards were to be implemented in a period of
less than six months to a year, manufacturers would be forced to write off old inventory that they
could no longer sell.

2-81

2.12.9 Efficiency Levels


One manufacturer was concerned that if the rulemaking process took too much time to
complete, new standards could be lower than efficiency levels the industry had already reached.
2.12.10

Regulatory Approval

Some manufacturers mentioned that regulatory approval for new products can come at a
significant cost while also requiring a substantial amount of time.
Related to standards for battery chargers, manufacturers and other parties identified three
general areas of concern: (a) conversion costs, (b) efficiency as a selling point, and (c) profit
margins.
2.12.11

Conversion Costs

Manufacturers stated that significant cost and time is required for testing and obtaining
safety approval for new products. Especially stringent standards would require significant capital
conversion costs.
2.12.12

Efficiency as a Selling Point

Some manufacturers commented that energy efficiency is not an important consideration


for buyers of battery chargers. The important factors to consumers are price and quality.
2.12.13

Profit Margins

Multiple manufacturers stated that new standards could raise product costs and hurt
manufacturers profit margins. These manufacturers asserted that they have been unable to pass
on increases in product costs to consumers in the past.
2.13

UTILITY IMPACT ANALYSIS

The utility impact analysis includes an analysis of the effects of new energy conservation
standards on the electric utility industry. For this analysis, DOE adapted NEMS, which is a large,
multi-sectoral, partial-equilibrium model of the U.S. energy sector that the EIA has developed
throughout the past decade, primarily for preparing EIAs Annual Energy Outlook (AEO). In
previous rulemakings, a variant of NEMS (currently termed NEMS-BT, BT referring to DOEs
Building Technologies Program), was developed to better address the specific impacts of an
energy conservation standard. NEMS, which is available in the public domain, produces a widely
recognized baseline energy forecast for the United States through the year 2035. The typical
NEMS outputs include forecasts of electricity sales, prices, and electric generating capacity.
DOE typically conducts the utility impact analysis as a scenario that departs from the
latest AEO reference case. In other words, the energy savings impacts from amended energy
conservation standards are modeled using NEMS-BT to generate forecasts that deviate from the
AEO reference case.

2-82

2.14

EMPLOYMENT IMPACT ANALYSIS

The imposition of standards can affect employment both directly and indirectly. Direct
employment impacts are changes, produced by new standards, in the number of employees at
plants that produce the covered products and at the affiliated distribution and service companies.
DOE evaluated direct employment impacts in the manufacturer impact analysis. Indirect
employment impacts that occur because of the imposition of standards may result from
consumers shifting expenditures between goods (the substitution effect) and from changes in
income and overall expenditure levels (the income effect).
DOE plans to utilize Pacific Northwest National Laboratorys impact of sector energy
technologies (ImSET) model to investigate the combined direct and indirect employment
impacts. The ImSET model, which was developed for DOEs Office of Planning, Budget, and
Analysis, estimates the employment and income effects energy-saving technologies produced in
buildings, industry, and transportation. In comparison with simple economic multiplier
approaches, ImSET allows for more complete and automated analysis of the economic impacts
of energy conservation investments.
2.15

ENVIRONMENTAL ASSESSMENT

Pursuant to the National Environmental Policy Act of 1969 (NEPA) (42 U.S.C. 4321 et
seq.) 42 U.S.C. 6295(o)(2)(B)(i)(VI), DOE intends to prepare a assessment (EA) of the potential
impacts that will quantify and consider the environmental effects of new or amended energy
conservation standards for battery chargers and external power supplies. The primary
environmental effects of these standards would be reduced power plant emissions resulting from
reduced consumption of electricity. DOE will assess these environmental effects by using
NEMS-BT to provide key inputs to its analysis. The portion of the environmental assessment
that will be produced by NEMS-BT considers carbon dioxide (CO2), nitrogen oxides (NOX), and
mercury (Hg). The environmental assessment also considers impacts on SO2 emissions. After a
brief discussion of general methodology, this section will address each of the relevant emissions.
This section then explains how DOE plans to monetize the benefits associated with emissions
reductions.

2.15.1 Carbon Dioxide


In the absence of any Federal emissions control regulation of power plant emissions of
CO2, a DOE standard is likely to result in reductions of these emissions. The CO2 emission
reductions likely to result from a standard will be estimated using NEMS-BT and national energy
savings estimates drawn from the NIA spreadsheet model. The net benefit of the standard is the
difference between emissions estimated by NEMS-BT at each standard level considered and the
AEO Reference Case. NEMS-BT tracks CO2 emissions using a detailed module that provides
results with broad coverage of all sectors and inclusion of interactive effects.

2-83

2.15.2 Sulfur Dioxide


DOE has determined that SO2 emissions from affected Electric Generating Units (EGUs)
are subject to nationwide and regional emissions cap and trading programs that create
uncertainty about the standards impact on SO2 emissions. The costs of meeting such emission
cap requirements are reflected in the electricity prices and forecasts used in DOEs analysis of
the standards. Title IV of the Clean Air Act sets an annual emissions cap on SO2 for all affected
EGUs. SO2 emissions from 28 eastern States and the District of Columbia (DC) are also limited
under the Clean Air Interstate Rule (CAIR, published in the Federal Register on May 12, 2005.
70 FR 25162 (May 12, 2005)), which creates an allowance-based trading program that will
gradually replace the Title IV program in those States and DC. (The recent legal history
surrounding CAIR is discussed below.) The attainment of the emissions caps is flexible among
EGUs and is enforced through the use of emissions allowances and tradable permits. Under
existing EPA regulations, any excess SO2 emission allowances resulting from the lower
electricity demand caused by the imposition of an efficiency standard could be used to permit
offsetting increases in SO2 emissions by any regulated EGU. However, if the standard resulted in
a permanent increase in the quantity of unused emission allowances, there would be an overall
reduction in SO2 emissions from the standards. While there remains some uncertainty about the
ultimate effects of efficiency standards on SO2 emissions covered by the existing cap and trade
system, the NEMS-BT modeling system that DOE plans to use to forecast emissions reductions
currently indicates that no physical reductions in power sector emissions would occur for SO2.
2.15.3 Nitrogen Oxides
NEMS-BT also has an algorithm for estimating NOX emissions from power generation.
The impact of these emissions, however, will be affected by the CAIR, which the Environmental
Protection Agency (EPA) issued on May 12, 2005. CAIR will permanently cap emissions of
NOX in 28 eastern states and the District of Columbia. 70 FR 25162 (May 12, 2005).
Much like SO2 emissions, a cap on NOX emissions means that battery charger and
external power supply standards may have little or no physical effect on these emissions in the
region covered by CAIR. Although CAIR has been remanded to the EPA by the DC Circuit, it
will remain in effect until it is replaced by a rule consistent with the Courts July 11, 2008,
opinion in North Carolina v. EPA. 531 F.3d 896 (DC Cir. 2008); see also North Carolina v.
EPA, 550 F.3d 1176 (DC Cir. 2008). Because all States covered by CAIR opted to reduce NOX
emissions through participation in cap-and-trade programs for electric generating units,
emissions from these sources are capped across the CAIR region.
DOE plans to use NEMS-BT to estimate the emissions reductions from possible
standards in the 22 States where emissions are not capped.
2.15.4 Mercury
Similar to emissions of SO2 and NOX, future emissions of Hg would have been subject to
emissions caps. In May 2005, EPA issued the Clean Air Mercury Rule (CAMR). 70 FR 28606
(May 18, 2005). CAMR would have permanently capped emissions of mercury for new and
existing coal-fired power plants in all States by 2010. However, on February 8, 2008, the DC

2-84

Circuit issued its decision in New Jersey v. Environmental Protection Agency, in which the DC
Circuit, among other actions, vacated the CAMR. 517 F.3d 574 (DC Cir. 2008). EPA has
decided to develop emissions standards for power plants under the Clean Air Act (Section 112),
consistent with the DC Circuits opinion on the CAMR. See
http://www.epa.gov/air/mercuryrule/pdfs/certpetition_withdrawal.pdf. Pending EPA's
forthcoming revisions to the rule, DOE is excluding the CAMR from its environmental analysis.
In the absence of CAMR, a DOE standard would likely reduce Hg emissions and DOE plans to
use NEMS-BT to estimate these emission reductions. However, DOE continues to review the
impact of rules that reduce energy consumption on Hg emissions and may revise its assessment
of Hg emissions reductions in future rulemakings.
2.15.5 Particulate Matter
Earthjustice commented that the significant contribution of power plants to [particulate
matter (PM)] pollution requires DOE to evaluate the impact of efficiency standards on PM
emissions. To support this statement, Earthjustice elaborated that power plants emit 7.2% of all
anthropogenic PM10 and 20.4% of all anthropogenic PM2.5. (Earthjustice, No. 18 at p. 3)
DOE acknowledges that particulate matter (PM) impacts are of concern due to human
exposures that can impact health. However, impacts of PM emissions reduction are much more
difficult to estimate than other emissions reductions due to the complex interactions between
PM, other power plant emissions, meteorology, and atmospheric chemistry that affect human
exposure to particulates. Human exposure to PM usually occurs at a significant distance from the
power plants that are emitting particulates and particulate precursors. When power plant
emissions travel this distance, they undergo highly complex atmospheric chemical reactions.
Although the EPA does keep inventories of direct PM emissions of power plants, in its source
attribution reviews the EPA does not separate direct PM emissions from power plants from the
sulfate particulates indirectly produced through complex atmospheric chemical reactions. The
great majority of PM emissions from power plants are of these secondary particles (secondary
sulfates). Thus, it is not useful to examine how new or amended standards impact direct PM
emissions independent of indirect PM production and atmospheric dynamics. Therefore, DOE is
not planning to assess the impact of these standards on particulate emissions. Further, even the
cumulative impact of PM emissions from power plants and indirect emissions of pollutants from
other sources is unlikely to be significant.
2.15.6 Monetization of Emissions Reduction Benefits
For those emissions for which real national emission reductions are anticipated (CO2, Hg,
and NOX for 22 states), only ranges of estimated economic values based on environmental
damage studies of varying quality and applicability are available. Therefore, DOE intends to
report estimates of monetary benefits derived using these values and consider these benefits in
weighing the costs and benefits of each of the standard levels considered.
In order to estimate the monetary value of benefits resulting from reduced emissions of
CO2 emissions, it is DOEs intent to use in its analysis the most current Social Cost of Carbon
(SCC) values developed and/or agreed to by interagency reviews. The SCC is intended to be a
monetary measure of the incremental damage resulting from greenhouse gas (GHG) emissions,

2-85

including, but not limited to, net agricultural productivity loss, human health effects, property
damage from sea level rise, and changes in ecosystem services. Any effort to quantify and to
monetize the harms associated with climate change will raise serious questions of science,
economics, and ethics. But with full regard for the limits of both quantification and monetization,
the SCC can be used to provide estimates of the social benefits of reductions in GHG emissions.
At the time of this notice, the most recent interagency estimates of the potential global
benefits resulting from reduced CO2 emissions in 2010 were $4.70, $21.40, $35.10, and $64.90
per metric ton in 2007 dollars. These values are then adjusted to 2009 dollars using the standard
GDP deflator value for 2008 and 2009. For emissions (or emission reductions) that occur in later
years, these values grow in real terms over time. Additionally, the interagency group determined
that a range of values from 7 percent to 23 percent should be used to adjust the global SCC to
calculate domestic effects, although preference will be given to consideration of the global
benefits of reducing CO2 emissions. See appendix 15A of the TSD preliminary analysis for the
full range of annual SCC estimates from 2010 to 2050. To calculate a present value of the stream
of monetary values, DOE will discount the values in each of the four cases using the discount
rates that had been used to obtain the SCC values in each case.
DOE recognizes that scientific and economic knowledge continues to evolve rapidly as to
the contribution of CO2 and other GHG to changes in the future global climate and the potential
resulting damages to the world economy. Thus, these values are subject to change.
DOE also intends to estimate the potential monetary benefit of reduced NOX emissions
resulting from the standard levels it considers. For NOX emissions, available estimates suggest a
very wide range of monetary values, ranging from $370 per ton to $3,800 per ton of NOX from
stationary sources, measured in 2001$ (equivalent to a range of $447 to $4,591 per ton in
2009$). Refer to the OMB, Office of Information and Regulatory Affairs, 2006 Report to
Congress on the Costs and Benefits of Federal Regulations and Unfunded Mandates on State,
Local, and Tribal Entities, for additional information. In accordance with U.S. Office of
Management and Budget (OMB) guidance, DOE will conduct two calculations of the monetary
benefits derived using each of the economic values used for NOX, one using a real discount rate
of 3 percent and another using a real discount rate of 7 percent.j
DOE does not plan to monetize estimates of Hg in this rulemaking. DOE is aware of
multiple agency efforts to determine the appropriate range of values used in evaluating the
potential economic benefits of reduced Hg emissions. DOE has decided to await further guidance
regarding consistent valuation and reporting of Hg emissions before it once again monetizes Hg
in its rulemakings.
2.16

REGULATORY IMPACT ANALYSIS

In the NOPR stage, DOE will prepare a regulatory impact analysis (RIA) pursuant to
Executive Order 12866, Regulatory Planning and Review, 58 FR 51735, October 4, 1993, which
is subject to review under the Executive Order by the Office of Information and Regulatory
Affairs at the Office of Management and Budget. The RIA addresses the potential for nonj

OMB, Circular A-4: Regulatory Analysis (Sept. 17, 2003).

2-86

regulatory approaches to supplant or augment energy conservation standards in order to improve


the energy efficiency or reduce the energy consumption of the products covered under this
rulemaking.
DOE recognizes that voluntary or other non-regulatory efforts by manufacturers, utilities,
and other interested parties can substantially affect energy efficiency or reduce energy
consumption. DOE plans to base its assessment on the actual impacts of any such initiatives to
date, but also will also consider, to the extent possible, information presented by interested
parties regarding the impacts existing initiatives might have in the future.
2.17

ADDITIONAL COMMENTS FROM INTERESTED PARTIES

DOE received several additional comments it wishes to acknowledge here that do not
pertain directly to the preliminary analyses.
2.17.1 Temporary Exemptions from Standards for Replacement Parts
An ITI member company requested that DOE include an exemption from test and
regulation for service parts (battery packs & external power supplies) intended for systems
manufactured prior to any regulatory effective date. (ITI member, No. 17 at p. 1)
According to section 325 of EPCA, as modified by EISA 2007, a Class A EPS is
excluded from standards if it is:
(i) manufactured during the period beginning on July 1, 2008, and ending on June 30,
2015; and
(ii) made available by the manufacturer as a service part or a spare part for an end-use
product
(I) that constitutes the primary load; and
(II) was manufactured before July 1, 2008 (42 U.S.C. 6295(u)(3)(B)).
The statutory language supports the ITI members request by excluding any EPS service
part for an end-use product manufactured before July 2008. This exclusion allows manufacturers
to make EPS spare parts until 2015 for end-use products manufactured prior to the compliance
date of the Class A standard. There are no similar exclusions for non-Class A EPSs or BCs.
2.17.2 Labeling Requirements and Marking Protocol
An ITI member company requested that DOE not require any international efficiency
markings, e.g., IV or V, on battery charger systems. (ITI member, No. 17 at p. 1) DOE may
prescribe labeling requirements for any covered product for which a test procedure has been
prescribed. (42 U.S.C. 6294(a)(5)) While DOE has not prescribed labeling requirements for
BCs or non-Class A EPSs, it reserves the right to do so at a later date, particularly if labeling
would aid in the identification of compliant products.

2-87

2.17.3 Standards Compliance and Enforcement


Vivitar pointed out that Federal government agencies regulating the toy, leather, and
textile industries have sought the assistance of test laboratories in their efforts to boost regulatory
compliance and suggested that DOE do the same. (Vivitar, Pub. Mtg. Tr., No. 14 at p. 377) DOE
appreciates this comment and will consider educating test laboratories on how to facilitate
compliance with BC and EPS standards as part of its broader compliance and enforcement
efforts.
2.17.4 Publishing Public Comments on the Internet
ASAP and Earthjustice both requested that public comments be made available online.
(ASAP, Pub. Mtg. Tr., No. 14 at p. 39; Earthjustice, Pub. Mtg. Tr., No. 14 at p. 47) DOE is
exploring options for doing so.
2.17.5 Public Meeting Webcast
ASAP expressed support for DOEs webcasting of the framework public meeting, noting
that this type of effort improves public accessibility to the rulemaking process accessible to the
public. (ASAP, Pub. Mtg. Tr., No. 14 at p. 39) DOE will continue to make efforts to ensure
public participation in its rulemaking process.

2-88

REFERENCES
1

Pacific Gas and Electric Company, Proposal Information Template for: Battery Charger Systems, submitted to
the California Energy Commission, 2008 Rulemaking Proceeding on Appliance Efficiency Regulations, Docket
Number 07-AAER-3, 2008
2
Taken in September 2009 from the ENERGY STAR 2.0 database at
http://www.energystar.gov/index.cfm?fuseaction=find_a_product.showProductGroup&pgw_code=ACD
3
Commission Regulation (EC) No. 278/2009 of April 6, 2009 implementing Directive 2005/32/EC of the European
Parliament and of the Council with regard to ecodesign requirements for no-load condition electric power
consumption and average active efficiency of external power supplies.

2-89

CHAPTER 3. MARKET AND TECHNOLOGY ASSESSMENT

TABLE OF CONTENTS
3.1

3.2

INTRODUCTION ........................................................................................................... 3-1

3.1.1 Definitions and Nomenclature ............................................................................. 3-1

3.1.1.1 Current Statutory Definitions................................................................... 3-1

3.1.1.2 Additional Definitions ............................................................................. 3-1

Wall Adapter............................................................................................ 3-2

Battery3-3

Battery Charging System ......................................................................... 3-3

3.1.2 Test Procedures.................................................................................................... 3-3

MARKET ASSESSMENT .............................................................................................. 3-6

3.2.1 Applications that Use BCs and EPSs................................................................... 3-6

3.2.1.1 Audio 3-7

MP3 Players ............................................................................................. 3-8

MP3 Speaker Docks and Clock Radios ................................................... 3-8

Wireless Speakers .................................................................................... 3-8

3.2.1.2 Computers and Peripherals ...................................................................... 3-9

Computers ................................................................................................ 3-9

Desktop Accessories .............................................................................. 3-10

Document Manipulation ........................................................................ 3-10

Document Readers ................................................................................. 3-11

Networking ............................................................................................ 3-11

3.2.1.3 Geospatial Equipment............................................................................ 3-11

3.2.1.4 Telephony .............................................................................................. 3-12

Mobile Telephony.................................................................................. 3-13

GSMA Universal Charging Solution. .................................................... 3-13

Stationary Telephony ............................................................................. 3-14

3.2.1.5 Household .............................................................................................. 3-15

Photo/Video ........................................................................................... 3-16

Floor Care .............................................................................................. 3-18

Kitchen Appliances................................................................................ 3-18

Childcare ................................................................................................ 3-19

Entertainment......................................................................................... 3-19

Home Systems ....................................................................................... 3-20

Other Household .................................................................................... 3-21

3.2.1.6 Outdoor Appliances ............................................................................... 3-21

3.2.1.7 Personal Care ......................................................................................... 3-22

3.2.1.8 Power Tools ........................................................................................... 3-24

3.2.1.9 Transport ................................................................................................ 3-25

Electric Vehicles .................................................................................... 3-26

Mobility Devices.................................................................................... 3-27

3.2.2 Shipments, Lifetimes, and Energy Performance................................................ 3-27

3-i

3.2.2.1 External Power Supply Shipments, Lifetimes and Energy Performance .. 3


28

3.2.2.2 Battery Charger Shipments, Lifetimes and Energy Performance .......... 3-36

3.2.3 Other Energy Efficiency Programs .................................................................... 3-40

3.2.4 Production and Distribution............................................................................... 3-43

3.2.5 Small Businesses................................................................................................ 3-45

3.2.6 Manufacturers and Market Shares ..................................................................... 3-45

3.2.7 Trade Associations and Other Interested Parties ............................................... 3-46

3.3
PRODUCT CLASSES................................................................................................... 3-49

3.3.1 EPS Product Classes .......................................................................................... 3-49

3.3.1.1 Nameplate Output Power ....................................................................... 3-49

3.3.1.2 Nameplate Output Voltage .................................................................... 3-51

3.3.1.3 Type of Power Conversion (AC/AC versus AC/DC) ............................ 3-52

3.3.1.4 Use with Medical Equipment................................................................. 3-52

3.3.1.5 Use with Battery Chargers of Motorized Applications and Detachable

Batteries ................................................................................................. 3-53

3.3.1.6 EPS Product Classes Under Consideration............................................ 3-53

3.3.2 BC Product Classes............................................................................................ 3-54

3.3.2.1 BC Product Class Criteria Considered for the Preliminary Analysis .... 3-54

3.3.2.2 Impacts of Topology on Product Class Selection .................................. 3-56

3.3.2.3 Resultant BC Product Classes................................................................ 3-57

3.4
TECHNOLOGY ASSESSMENT.................................................................................. 3-58

3.4.1 Introduction........................................................................................................ 3-58

3.4.2 EPS Modes of Operation ................................................................................... 3-58

3.4.2.1 EPS Active Mode................................................................................... 3-58

3.4.2.2 EPS No-Load Mode............................................................................... 3-60

3.4.3 EPS Efficiency Metrics...................................................................................... 3-60

3.4.4 Energy Efficiency Metrics for External Power Supplies................................... 3-60

3.4.5 EPS Designs....................................................................................................... 3-60

3.4.5.1 AC/AC External Power Supplies........................................................... 3-61

3.4.5.2 Unregulated Line-Frequency AC/DC External Power Supplies ........... 3-62

3.4.5.3 Linear-Regulated Line-Frequency AC/DC External Power Supplies ... 3-62

3.4.5.4 Switching-Regulated Line-Frequency AC/DC External Power Supplies . 3


65

3.4.5.5 Switched-Mode AC/DC External Power Supplies ................................ 3-66

3.4.6 EPS Technology Options................................................................................... 3-69

3.4.7 BC Modes of Operation ..................................................................................... 3-70

3.4.7.1 Active or Charge Mode.......................................................................... 3-70

3.4.7.2 Maintenance Mode................................................................................. 3-70

3.4.7.3 Standby or No-Battery Mode................................................................. 3-70

3.4.7.4 Other Modes and Applicability.............................................................. 3-71

3.4.8 BC Efficiency Metrics ....................................................................................... 3-71

3.4.9 Battery Charger Design...................................................................................... 3-72

3.4.10 Battery Charger Technology Options ................................................................ 3-74

REFERENCES .......................................................................................................................... 3-77

3-ii

LIST OF TABLES

TABLE 3.1 SHIPMENTS BY APPLICATION, THOUSANDS ............................................... 3-9

TABLE 3.2 ANNUAL TELEPHONE UNIT SHIPMENTS (THOUSANDS) AND CAGR ... 3-15

TABLE 3.3 EXTERNAL POWER SUPPLY LIFETIMES AND SHIPMENTS BY PRODUCT

CLASS ........................................................................................................................... 3-29

TABLE 3.4 EPS PRODUCT CLASS A1A: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-30

TABLE 3.5 EPS PRODUCT CLASS A1B: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-30

TABLE 3.6 EPS PRODUCT CLASS A1C: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-31

TABLE 3.7 EPS PRODUCT CLASS A1D: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-31

TABLE 3.8 EPS PRODUCT CLASS A2: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-31

TABLE 3.9 EPS PRODUCT CLASS A3: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-32

TABLE 3.10 EPS PRODUCT CLASS A4: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-32

TABLE 3.11 EPS PRODUCT CLASS X1: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-32

TABLE 3.12 EPS PRODUCT CLASS X2: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-32

TABLE 3.13 EPS PRODUCT CLASS H1: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-32

TABLE 3.14 EPS PRODUCT CLASS M1: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-33

TABLE 3.15 EPS PRODUCT CLASS M2: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-33

TABLE 3.16 EPS PRODUCT CLASS M3: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-33

TABLE 3.17 EPS PRODUCT CLASS M4: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-33

TABLE 3.18 EPS PRODUCT CLASS B1: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-34

TABLE 3.19 EPS PRODUCT CLASS B2: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-34

TABLE 3.20 EPS PRODUCT CLASS B3: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-34

TABLE 3.21 EPS PRODUCT CLASS B4: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-34

TABLE 3.22 ENERGY PERFORMANCE OF NEW EXTERNAL POWER SUPPLIES IN 2009

........................................................................................................................................ 3-35

TABLE 3.23 ENERGY PERFORMANCE OF EXTERNAL POWER SUPPLIES IN 2007 .. 3-36

3-iii

TABLE 3.24 BATTERY CHARGER LIFETIMES AND SHIPMENTS BY PRODUCT CLASS

........................................................................................................................................ 3-36

TABLE 3.25 BC PRODUCT CLASS 1: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-37

TABLE 3.26 BC PRODUCT CLASS 2: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-37

TABLE 3.27 BC PRODUCT CLASS 3: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-37

TABLE 3.28 BC PRODUCT CLASS 4: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-38

TABLE 3.29 BC PRODUCT CLASS 5: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-38

TABLE 3.30 BC PRODUCT CLASS 6: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-38

TABLE 3.31 BC PRODUCT CLASS 7: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-38

TABLE 3.32 BC PRODUCT CLASS 8: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-39

TABLE 3.33 BC PRODUCT CLASS 9: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-39

TABLE 3.34 BC PRODUCT CLASS 10: TOP APPLICATIONS, SHIPMENTS, AND

LIFETIMES ................................................................................................................... 3-39

TABLE 3.35 ENERGY PERFORMANCE OF NEW BATTERY CHARGERS IN 2009....... 3-40

TABLE 3.36 BC AND EPS EFFICIENCY PROGRAMS WORLDWIDE.............................. 3-42

TABLE 3.37 TRADE ASSOCIATIONS .................................................................................. 3-47

TABLE 3.38 COMPARISON OF I2 R LOSSES FOR TWO 20-WATT EPSS .................... 3-51

TABLE 3.40 PROPOSED CLASS A EPS PRODUCT CLASSES. ......................................... 3-53

TABLE 3.41 PROPOSED MULTIPLE VOLTAGE EPS PRODUCT CLASSES................... 3-53

TABLE 3.42 PROPOSED HIGH POWER EPS PRODUCT CLASSES.................................. 3-53

TABLE 3.43 PROPOSED MEDICAL EPS PRODUCT CLASSES. ....................................... 3-54

TABLE 3.44 PROPOSED MADB EPS PRODUCT CLASSES. ............................................. 3-54

TABLE 3.45 BC PRODUCT CLASSES USED IN THE PRELIMINARY ANALYSIS. ....... 3-58

LIST OF FIGURES
FIGURE 3.1 EXAMPLE OF A WALL ADAPTER AND A DESKTOP ADAPTER, TREATED

TOGETHER AS WALL ADAPTERS......................................................................... 3-2

FIGURE 3.2 EXAMPLE OF A CRADLE AND A CHARGING BASE ................................... 3-3

FIGURE 3.3 EXAMPLES OF BATTERIES .............................................................................. 3-3

FIGURE 3.4 PATHS OF DISTRIBUTION FOR BATTERY CHARGERS AND EXTERNAL

POWER SUPPLIES....................................................................................................... 3-44

FIGURE 3.5 TOP EPS MANUFACTURERS IN 2005 ............................................................ 3-46

FIGURE 3.6 THE EISA STANDARD AND ENERGY STAR 2.0 SPECIFICATION FOR

AVERAGE EFFICIENCY ............................................................................................ 3-50

FIGURE 3.7 THE EISA STANDARD AND ENERGY STAR 2.0 SPECIFICATION FOR NO
LOAD POWER CONSUMPTION................................................................................ 3-50

3-iv

FIGURE 3-8 EPSS QUALIFYING FOR ENERGY STAR 2.0 AS OF SEPTEMBER 2009. . 3-52

FIGURE 3.9 EXAMPLE OF AN EFFICIENCY CURVE OF AN EPS IN ACTIVE MODE . 3-59

FIGURE 3.10 CIRCUIT DIAGRAM FOR AN AC/AC EXTERNAL POWER SUPPLY ...... 3-61

FIGURE 3.11 CIRCUIT DIAGRAM OF A LINE-FREQUENCY RAW SUPPLY ................ 3-62

FIGURE 3.12 BLOCK DIAGRAM OF A LINEAR REGULATOR........................................ 3-64

FIGURE 3.13 SIMPLIFIED CIRCUIT DIAGRAM OF A LINEAR REGULATOR .............. 3-64

FIGURE 3.14 BLOCK DIAGRAM OF A SWITCHING-REGULATED LINE-FREQUENCY

AC/DC EPS ................................................................................................................... 3-66

FIGURE 3.15 BLOCK DIAGRAM OF A SWITCHED-MODE POWER SUPPLY............... 3-67

FIGURE 3.16 SIMPLIFIED CIRCUIT DIAGRAM OF A FLYBACK SWITCHING

REGULATOR ............................................................................................................... 3-68

FIGURE 3.17 GENERAL SCHEMATIC OF A BC AND BATTERY.................................... 3-72

3-v

CHAPTER 3. MARKET AND TECHNOLOGY ASSESSMENT

3.1

3.1.1

INTRODUCTION

Definitions and Nomenclature

A key issue in DOEs BC and EPS rulemakings is the necessity for clear and distinct
product definitions. DOE has taken several approaches in the past, but with the current
rulemakings, the need for regulatory definitions based on an interpretation of the EPCA statutory
language is becoming more pressing. Therefore, before discussing any analysis, DOE will
review the current statutory definitions and discuss the possible interpretation of these
definitions, with the goal of establishing regulatory definitions that provide greater clarity for
interested parties.
3.1.1.1 Current Statutory Definitions
Title III of EPCA (42 U.S.C. 6291, et seq.) sets forth a variety of provisions designed to
improve energy efficiency. Part A of Title III (42 U.S.C. 62916309) establishes the Energy
Conservation Program for Consumer Products Other Than Automobiles. Section 135 of EPACT
2005, Pub. L. 109-58, amended sections 321 and 325 of EPCA, inserting definitions for BCs and
EPSs into the list of covered products. Subsequently, Section 301 of EISA created a subset of
EPSs designated as Class A. Chapter 2 of this preliminary TSD presents the statutory definitions
for:
Battery Charger
External Power Supply
Class A External Power Supply
Consumer Product
Detachable Battery

3.1.1.2 Additional Definitions


DOE developed additional definitions to clarify ambiguous terms in the rulemaking.
Specifically, determining which parts of a battery charging system are considered the battery
charger (BC), and therefore subject to this standards rulemaking, depends on the interpretation of
the BC definition, just as which wall adapters are considered EPSs depends on the interpretation
of the EPS definition. In other words, BC and EPS are regulatory terms subject to
interpretation, while battery charging system and wall adapter are terms that describe
physical objects. Throughout the preliminary analysis, DOE has taken care to distinguish
between these regulatory and physical terms, defined as follows.

3-1

Wall Adapter
DOE uses the term wall adapter to refer to any external power adapter (Figure 3.1) that
connects a consumer product to the household electric supply (AC mains), regardless of whether
the device contains charge control circuitry. Wall adapters, which consist of a power conversion
circuit housed in a plastic enclosure and typically connected to a product through an output cord,
provide many functions, the most important of which is safety. All wall adapters isolate the
consumer product from mains and reduce the voltage, thereby reducing the risk of shock.
In addition to performing isolation and conversion functions, wall adapters may also
provide functions such as rectification (AC/DC conversion), voltage regulation, and/or control of
the charge current from AC mains to the battery for safe charging. There are no distinguishing
physical features (e.g., size, shape, etc.) that would definitively allow an observer to determine
the internal circuitry (and the corresponding functions) of a wall adapter, although certain
physical features have been used in other programs. Furthermore, some adapters can provide
different functions depending on their region of operationi.e., the output current and voltage at
a particular time. Appendix 3C presents an analysis of criteria that could be used to determine
the presence of charge control in a wall adapter.

Wall Adapter

Desktop Adapter

Figure 3.1 Example of a Wall Adapter and a Desktop Adapter, Treated Together as Wall
Adapters
Cradle or Charging Base
For many rechargeable consumer products, the battery may be charged using a cradle
or charging base (Figure 2.2). Some cradles and charging bases use a wall adapter in tandem,
while others perform the same power conversion and isolation functions, rendering wall adapters
unnecessary. Similar to wall adapters, cradles and charging bases have no distinguishing physical
features that would allow an observer to conclusively determine their internal circuitry and
function.

3-2

Handheld vacuum cleaner cradle


(shown with wall adapter)

Power tool battery charging base


(no wall adapter)

Figure 3.2 Example of a Cradle and a Charging Base

Battery
For this rulemaking, DOE considers batteries to be one or more sealed electrochemical
cells that provide power to a consumer product, allowing it to operate while disconnected from
AC mains. Products may use batteries in standard-size (e.g., AA, AAA, etc.) packages or non
standard, product-specific packages. Batteries may also be packaged with additional circuitry to
prevent overcharging, or to detect faults or charge status. Examples of batteries are illustrated in
Figure 3.3.

Laptop battery
pack

Standard-size batteries

Power tool battery pack


shown with sub-C cell (not to scale)

Figure 3.3 Examples of Batteries

Battery Charging System


The term battery charging system refers to all the components necessary to charge and
maintain a battery, from the AC wall plug to the battery itself, and includes the wall adapter,
cradle, andif the battery is integral to the consumer productthe product itself.

3.1.2

Test Procedures

Section 323 of EPCA (42 U.S.C. 6293) sets forth generally applicable criteria and
procedures for DOEs adoption and amendment of test procedures, which manufacturers of
covered products must use to certify to the U.S. Department of Energy (DOE) that their products
comply with EPCA energy conservation standards and to quantify the efficiency of their

3-3

products. Also, these test procedures must be used whenever testing is required in an
enforcement action to determine whether covered products comply with EPCA standards.
DOE has adopted test procedures for both BCs and EPSs. These are described in turn in
the sections below, while further discussion of testing and efficiency metrics and their
application to the analysis of achievable performance can be found later sections 3.4.3 for EPSs
and 3.4.8 for BCs.
Within this document, the term energy consumption is used loosely to refer to power
dissipation of BCs and EPSs in one of their several modes, taking into account the time spent in
that mode. Likewise, while efficiency has a precise definition,a it can also refer more generally
to the energy consumption of a BC or EPS in any of their modes when compared to similar
devices. Thus devices with lower energy consumption are referred to as more efficient, and vice
versa throughout this discussion.
3.1.2.1 EPS Test Procedures
On December 8, 2006, DOE codified a test procedure final rule for EPSs in appendix Z to
subpart B of 10 CFR Part 430 (Uniform Test Method for Measuring the Energy Consumption
of External Power Supplies). 71 FR 71340. DOEs test procedure, based on the ENERGY
STAR EPS test procedure, measures active mode efficiency and no-load mode (standby mode)
power consumption. In the standby and off mode test procedure NOPR for BCs and EPSs, 73 FR
48054 (August 15, 2008), DOE proposed to amend the EPS test procedure to add a measurement
of power consumption in off mode, where, if the EPS has an on-off switch, the EPS is connected
only to mains and the switch is turned off. These amendments were included in the final rule,
published March 27, 2009. 74 FR 13335. DOE is considering amending the EPS test procedure
as part of its revision to the BC test procedure NOPR published on April 2, 2010. 75 FR 16958.
Active mode conversion efficiency is the ratio of output power to input power. DOE
averages the efficiency at four loading conditions25, 50, 75, and 100 percent of maximum
rated output currentto assess the performance of an EPS. DOE also measures the power
consumption of the EPS when disconnected from the consumer product, which is termed no-load
power consumption. If the EPS has an on-off switch, the switch is on when conducting the
measurement.
3.1.2.2 BC Test Procedures
On December 8, 2006, DOE adopted a test method to measure the efficiency of battery
chargers. 71 FR 71340. This test method, based on the U.S. Environmental Protection Agencys
(EPA) ENERGY STAR Test Methodology for Determining the Energy Performance of Battery
Charging Systems, measures the power consumed by BCs in maintenance and no-battery, as
well as the energy recovered from the battery during discharge, calculating an energy ratio.
In the December 8, 2006, Test Procedure Final Rule, DOE stated that it intended to study
further BC active mode energy consumption and reserved a section in the test procedure (section
a

See, for instance, section 4.3.1 of IEEE standard 1515 or the CEC single-voltage EPS test procedure, incorporated
by reference into DOEs single-voltage EPS test procedure in appendix Z.

3-4

4(b) of appendix Y to subpart B of 10 CFR Part 430). 71 FR 71340, 71360. During the
framework document public meeting, DOE presented several potential approaches to measuring
battery charger energy consumption in active mode. Numerous interested parties submitted
comments on active mode measurement in general, and specifically on these sub-issues:
Harmonization of DOEs BC active mode test procedure with the BC test procedure
adopted by the California Energy Commission (CEC)
Limiting testing to U.S. line-voltage AC input (115 volts at 60 hertz)
Shortening the charge and maintenance mode testing period to less than 24 hours
Incorporating usage profiles into the test procedure
Measuring charger output energy instead of battery output energy
Measurements at alternative depths-of-discharge to better reflect consumer use
Test procedure rulemaking schedule
Appropriate voltage for testing multiple-voltage BCs
Power factor measurement
DOE took the comments on the above topics under advisement when developing
amendments to its existing BC test procedure to accommodate energy consumption in active
mode. Comments relevant to the proposal, as well as the proposed amendments themselves, are
discussed in depth in the BC active mode test procedure NOPR, published on April 2, 2010. 75
FR 16958.
If adopted, this proposed amendment would stop use of the non-active energy ratio in
favor of metrics related to energy consumption in each of the energy-consuming modes of
operation of a BC. As described above, these include active, maintenance, standby, and off
modes; thus, the test procedure would return four separate metrics. How these four modespecific metrics would be combined for the purpose of an energy conservation standard is
discussed further in chapter 5.

3-5

3.2

MARKET ASSESSMENT

3.2.1

Applications that Use BCs and EPSs

To characterize the market for BCs and EPSs, DOE gathered information on the products
that use them. DOE refers to these products as end-use consumer products or BC and EPS
applications. This method was chosen for two reasons. First, the demand for applications
drives the demand for BCs and EPSs because BCs and EPSs are nearly always integrated into,
bundled with, or otherwise intended to be used with a given application. Second, because most
BCs and EPSs are not stand-alone products, their usage profiles, energy consumption, and power
requirements are all determined by the associated application. Therefore, to develop reliable
estimates of the real-world unit energy consumption of a BC or EPS, it is necessary to examine
the application.
DOE began the development of the preliminary analysis by analyzing online and brickand-mortar retail outlets to determine which applications use BCs and EPSs and which BC and
EPS technologies are most prevalent. Non-Class A EPSs were analyzed as part of the non-Class
A EPS determination analysis, which was completed shortly before this market survey. NonClass A EPSs were, therefore, not addressed in this market survey; however, DOE will
incorporate them into an updated market survey at the NOPR stage of the standards rulemaking.
To best characterize the markets for BCs and EPSs, DOE focused its search on those applications
likely to have the greatest significance in the standards analyses (based on shipments, lifetimes,
and energy use). The survey consisted of the following steps:
1. Identified all applications that use BCs and Class A EPSs
2. Estimated annual shipments, lifetimes, and energy consumption for those applications.
3. Selected applications likely to have a significant impact on energy use based on estimated
shipments, lifetimes, and energy consumption.
4. For selected applications, visited websites and retail outlets to identify popular models
and document BC and EPS characteristics.
DOE then used this survey to select representative units and common BCs and EPSs to
be tested. This process is described in chapter 5. The results of this product survey are presented
in the Excel file BCEPS_Master_Survey.xls.
DOE has identified four major trends that can affect shipments of BCs and EPSs over
time. These trends are all related to the consumer products powered by BCs and EPSs.
Demand for Consumer Product Applications refers to the changes in preferences,
level of affluence, and population size that affect the demand for existing consumer
product applications that use BCs or EPSs.
Convergence means the application that uses an EPS is made redundant by another
application. For example, mobile telephones increasingly incorporate the features of
personal digital assistants (PDAs), digital cameras, portable media players, and
portable navigation devices. As a result of convergence, some consumers may now
have fewer devices than in the past, thus reducing the demand for BCs or EPSs.

3-6

Emergence refers to the creation of new consumer product application categoriesa


critical factor, given the rapid pace of change in the consumer electronics market.
Substitution means a shift between methods for supplying power to consumer
productsinternal power supplies, external power supplies, primary batteries,
rechargeable batteries, Universal Serial Bus (USB) systems, and others.
DOE identified approximately eighty applications that use BCs and EPSs. Although there
are certainly some BC and EPS applications that DOE did not consider, DOE believes it has
captured the vast majority of BC and EPS shipments for use with consumer products. Because
DOEs scope does not include BCs and EPSs used only in a commercial setting, DOE did not
estimate shipments of BCs and EPSs used with commercial products. For further discussion of
the scope of the BCs and EPSs covered in the preliminary analysis, please refer to chapter 2.
For ease of exposition, DOE grouped applications into nine categories. A categorized list
of applications is shown in Appendix 3A. For each category, the market assessment examines
major applications, shipments, lifetimes, and BC and EPS technical characteristics. Trends and
factors that may affect future shipments of BCs and EPSs are also discussed. Generally,
characteristics about the batteries, BCs, and EPSs used with each application were derived from
an extensive survey of products available at online retailers and in stores. DOE surveyed nearly
1,000 products to gather specific BC and EPS data (such as output voltage). The details of the
survey can be found in the Excel file BCEPS_Master_Survey.xls.
3.2.1.1 Audio
The audio equipment category includes both niche applications, such as guitar effects
pedals, and very common applications such as MP3 players. This category does not include
computer speakers. DOE estimates total shipments were 66 million in 2008. The most numerous
units include MP3 players, MP3 speaker docks, and clock radios; approximately 62 million units
shipped in 2008.1 DOE estimates that 20 percent of units use an EPS and 73 percent use a BC.
The prevalence of BC- or EPS-powered musical instruments and component audio equipment is
low; DOE estimates total BC shipments within these products to be less than 250,000 units and
EPS shipments to be approximately two million. DOE examined amateur radios as part of its
determination for non-Class A EPSs. DOE estimated annual shipments of 3,000 high power
EPSs for amateur radios. These EPSs typically have nameplate output powers of 345 watts. 74
FR 56928.
Amateur Radios
Component Audio

Pre-Amps
Wireless Speakers

Audio

Musical
Instruments

Guitar Effects Pedals


Keyboards
MP3 Speaker Docks

Portable Audio &


Accessories

Clock Radios
MP3 Players

3-7

MP3 Players
Portable media players such as MP3 players constitute the majority of shipments in the
audio category; 43.7 million units were shipped in 2008.1 The vast majority of these units
employ a BC. While shipments are high, CEA noted an 8.9 percent drop in shipments between
2007 and 2008. They attribute this decline to convergence with smart phones. In contrast, trends
to add additional features to portable media players, such as video, could increase demand for
these devices. CEA expects MP3 player market saturation to increase from 51 percent of
households in 2009 to 60 percent by 2012. The Pacific Gas and Electric Company and others
commented that lifetimes are estimated to be four years. (PG&E et al., No. 20 at p. 10) All
portable media players analyzed by DOE were powered by 3.7 volt batteries. Apple is the market
leader for portable media players with over 70 percent market share in 2007.2 b Battery energy
for Apple products ranges from 0.9 to 3.3 watt-hours. Overall, portable media players with color
display screens tend to use similar batteries as mobile phones. Nearly all portable media players
are recharged via USB connections, although many manufacturers offer EPSs that output voltage
at USB levels (five volts) as an optional accessory. DOE assumes that ten percent users that
purchase an MP3 player also purchase an aftermarket EPS.

MP3 Speaker Docks and Clock Radios


From researching common units for sale, DOE found that the majority of MP3 speaker
docks employ EPSs, while most traditional clock radios run directly from mains power.
However, convergence between these devices is increasing. In 2008, a total of 17.9 million clock
radios and MP3 speaker docks were shipped, and DOE estimates approximately 44 percent of
those units had EPSs.1 Of the models DOE examined, most used EPSs with nameplate output
power between 13 and 18 watts. DOE found a few models with EPSs as high as 60 watts of
output power. EPS output voltage clustered around 10, 12, and 15 volts.c DOE estimates that 15
percent of MP3 speaker docks contain integral rechargeable batteries and have BCs for those
batteries. DOE found battery information for only one such MP3 speaker dock. That model used
a 3.7 volt battery rated at 8 watt hours. Since most MP3 speaker docks can also charge the media
player, there is some question as to whether these docks contain BCs for this purpose. At present,
DOE believes that charge control lies within the media player while the MP3 speaker dock acts
simply as a power supply. DOE welcomes stakeholder comment on this issue. DOE assumes
lifetimes for MP3 speaker docks and clock radios to be 4 years.

Wireless Speakers
DOE estimates shipments of wireless speakers to be 750,000 units in 2008. This
application will likely experience strong growth as wireless technology improves and demand
increases. Wireless speakers require no cable to transmit audio output from the audio source, but
the wireless transmitter is powered by an EPS. Power to the speaker or speakers is provided by
one or two EPSs and DOE estimates that 1.5 million EPSs were shipped with wireless speakers
in 2008. DOE is aware of one wireless speaker model that is powered by a wall adapter and can
charge and operate on standard-sized C-cell rechargeable batteries.3

According to the NPD Group, the iPods market share was at 72.7 percent in January, 72.3 percent in February,

and 68.9 percent in March.

c
Based on wall adapter information from 20 MP3 speaker docks and clock radios.

3-8

3.2.1.2 Computers and Peripherals


This category includes all computers and related equipment. As this is a broad category,
applications have been separated into five subcategories: computers, desktop accessories,
document manipulation, document readers, and networking. These applications have lifetimes of
between 3 and 5 years. (PG&E et al., No. 20 at p. 10)4
Personal Digital Assistants
Computers

Netbooks
Notebooks
LED Monitors

Desktop
Accessories

Computer Speakers
External Hard Drives
Uninterruptible Power Supplies

Computers and Peripherals

Image Scanners
Document
Manipulation

Handheld Image Scanners


Inkjet Imaging Equipment
Portable Printers

Document Reader

Networking

E-Book Readers
LAN Equipment
Wi-Fi Access Points

Shipments totaled 118.6 million units in 2008 for the category as a whole. DOE estimates that
53.7 million BCs and nearly 88.7 million EPSs were shipped with these end-use products. DOE
estimated external hard drive shipments at over 600,000 units in 2008. Document manipulation
devices shipped 25.6 million units. Electronic book readers were the smallest subcategory in
terms of shipments, with 580,000 units shipped in 2008, though this segment is growing rapidly.1

Computers
BC/EPS shipments in the computers subcategory are high. For computers alone
(including notebooks, netbooks, and personal digital assistants (PDAs)), shipments were
approximately 46 million units in 2008.5 Applications covered in the computers subcategory are
built for portability and, as a result, use BCs and EPSs. PDAs have seen significant convergence
with mobile phones. In 2005, more than 4.7 million units were shipped, and that number
decreased to two million in 2008.1 The functionality of smart phones reduces the need for two
devices and, by extension, the need for multiple BCs. See Table 3.1 for an illustration of this
convergence. DOE includes smart phones in its analysis of mobile phones under the telephony
category.
Table 3.1 Shipments by Application, thousands
2005
2006
2007
PDAs
4,720
3,850
2,175
Smart
9,842
14,020
24,196
Phones

3-9

2008
1,977
39,884

CAGR (%)
-25
59

Based on its survey of the market, DOE found that all EPSs powering notebook and
netbook computers are similar in voltage (~20 V), but vary in output power due to differences in
intended functionality. Netbooks require wall adapter output powers of approximately 30-65 W,
while notebooks require 60-120 W. Both types typically use 11.1 V batteries. Battery energy is
similar between netbooks and notebooks, typically ranging from 40 to 60 watt hours. Due to
their smaller product and battery sizes, PDAs use wall adapters rated at outputs of 5 V to power
the device and/or recharge the devices 3.7 volt batteries. Although PDA shipments are in
decline, netbook shipments have grown significantly owing to their greater portability and lower
prices compared to full-sized notebooks.5

Desktop Accessories
Desktop accessories are applications designed for at-home use with personal computers.
Total shipments in 2008 were 17.8 million units and include computer speakers, external hard
drives, and uninterruptible power supplies. Without data on computer speaker shipments, DOE
assumed that speaker sales would be equivalent to sales of desktop computers, at 10.2 million
units in 2008.1 Based on its survey of products, DOE estimates that 38 percent of computer
speakers are powered by EPSs. Output power for these EPSs varied between 6 and 68 W. DOE
estimates that 58 percent of external hard drives, 374,000 units, used EPSs. 6,7,d Most units used
12 V EPSs with output power that varied between 12 and 57 W.
Uninterruptable power supplies (UPSs) contain BCs but do not use EPSs. UPSs act as
power strips with built-in batteries that remain charged in order to provide battery power to
attached devices in the event of a power surge or power interruption. Based on total revenue
from a report by Global Industry Analysts, DOE estimated annual sales of consumer UPSs by
examining the prevalence of UPSs at different capacity ratings available from online retailers.
DOE estimates annual shipments of consumer UPSs to be 6.9 million units.8 Most consumer
UPSs contain built-in 12 V batteries with energies ranging from 84 to 168 watt hours. Built-in
BCs are able to fully charge these batteries in 3 to 24 hours, though most can do so in between
4.5 and 16 hours.
DOE also identified an LED monitor that uses a 250 watt EPS, just meeting the Class A
definition. DOE estimates shipments were 160,000 units in 2008. It is possible that shipments of
high power non-class A EPSs will increase if future LED monitors move beyond 250 watts of
nameplate output power.

Document Manipulation
Document manipulation is another subcategory containing applications that are
experiencing significant convergence. DOE estimates total shipments in this subcategory to be
25.6 million units. Inkjet printers and multi-function devices (MFDs) make up the vast majority
of devices, at 23.3 million units.9 MFDs are inkjet printers that incorporate scanning, copying,
and faxing capabilities. The increase in the number of affordable MFD models has significantly
d

Worldwide shipments for external hard drives were 2.6 million in 2006. Based on the Darnell Groups estimated
distribution of computer shipments, North America makes up 29 percent of worldwide computer shipments. U.S.
GDP is 85 percent of North American GDP, yielding U.S. external hard drive shipments of 644,215. DOE estimates
that 58 percent of external hard drives use power supplies, resulting in EPS shipments of 373,644.

3-10

reduced the need for several distinct devices. Based on units available in the market, DOE found
that two major manufacturers, Hewlett Packard and Lexmark, make MFDs and inkjet printers
with EPSs, and DOE estimates they have a combined market share of 40 percent.10,11 Hence,
DOE estimates that 9.3 million used EPSs. Output power for most MFDs and printers varies
significantly within the range of 15-108 W, with most models ranging between 25 and 80 watts.
Output voltage was consistently at 30-32 V. In 2008 many Hewlett Packard MFDs used nonClass A multi-voltage EPSs. DOE estimates that 3.8 million non-Class A EPSs were shipped
with Hewlett Packard MFDs in 2008.

Document Readers
Electronic document readers, also known as e-book readers, are a quickly growing
subcategory. CEA predicted shipments would double between 2008 and 2009 to 1.2 million
units. These portable rechargeable devices enable users to download and display electronic
books. DOE surveyed five of these devices, four of which were conclusively found to use wall
adapters (with outputs of 4.2-10.4 W) to power their BCs. The Amazon Kindle, the original ebook reader, uses a 3.7 V battery with 5.7 watt hours of energy.

Networking
As wireless technologies gain market share over traditional modems, market saturation of
networking equipment will continue to increase. These devices include LAN equipment,
broadband modems, routers, and Wi-Fi access points. With 2008 shipments of 28.7 million units,
this is one of the largest subcategories in DOEs analysis.12 Of the units in this category analyzed
by DOE, 96 percent use EPSs and all of these provide output voltage between 5 V and 12 V,
with the majority outputting at the 12 V level. Most units have output powers around 12 watts.

3.2.1.3 Geospatial Equipment


The geospatial equipment category is comprised of rechargeable global positioning
system (GPS) devices, which include handheld and in-vehicle GPS devices. DOE believes that
in-vehicle GPS units make up the majority of the GPS market, given the rapid year over year
growth they have experienced. Shipments in 2008 were just under 15.3 million units, a
significant increase from the 8.8 million units shipped in 20071. Between 2005 and 2008, the
compound annual growth rate was 174 percent1. Since the majority of handheld GPS devices are
powered by primary batteries, DOE assumes that just 15 percent of the handheld devices sold are
rechargeable. A wall adapter for charging purposes is either included with the product or sold
separately. Shipment data is not readily available on handheld GPS units, but DOE estimates that
150,000 units shipped in 2008 use an EPS and BC. DOE assumes lifetimes for handheld and incar GPS units to be 5 years.

3-11

In-Vehicle GPS
Geospatial Equipment
Handheld GPS

In-vehicle GPS units are primarily charged by a DC car charger, though many can also be
charged with an AC-DC wall adapter. DOE assumes battery voltage for in-vehicle GPS units is
3.7 V with energy between 4 and 8 watt hours.
Convergence is gradually becoming a factor in this category as the market matures and
new functions are added to differentiate products and boost sales. Examples include GPS models
that double as two-way radios and GPS models with built in digital cameras. These features were
included in only a few of the products surveyed, so DOE does not anticipate these units to impact
BC and EPS shipments significantly. Many automobile manufacturers offer installed navigation
systems that have the potential to erode the long-term sales potential of stand-alone models.
Other substitutes include GPS-enabled smart phones. While these technologies are
limited by the availability of cellular service signals (in addition to GPS satellite data signals),
smart phones have the potential to limit GPS sales (for both handheld and in-car units). Due to
limits in mobile phone GPS functionality, however, DOE does not expect mobile phones to
completely replace GPS units. Regardless, this is an instance of convergence that could place
downward pressure on BC shipments.
3.2.1.4 Telephony
DOE has separated telephony into two sub-categories with very different power supply
and usage characteristics. These categories are mobile telephony (including two-way radios,
mobile/cellular telephones, and accessories for these devices) and stationary telephony
(including cordless telephones and satellite charging bases, caller ID devices, and voice over
internet protocol or VoIP adapters).
Bluetooth Headsets
Mobile

Consumer Two-Way Radios


Mobile Phones

Telephony

Caller ID Devices
Stationary

Cordless Phones
Answering Machines
VoIP Adapters

3-12

Shipments in this category vary widely, from standalone caller ID devices, of which
460,000 units shipped in 2008, to mobile phones, with shipments of 131.4 million units. Total
unit shipments for applications in this category were 189.8 million units. 1, 13

Mobile Telephony
Applications in the mobile telephony category are small, portable devices designed for
mobile communication. With the exception of those consumer two-way radios that are powered
by primary batteries, products in this subcategory use BCs. Mobile phones typically employ
EPSs, although substitution is a factor as some can also obtain power from USB ports. Bluetooth
headsets are typically charged with USB connectors or wall adapters.
Features of mobile phones and personal digital assistants (PDAs) have converged. Smart
phones incorporate the utility of handheld computers into mobile telephones, reducing the need
for two separate devices. CEA predicts that smart phones have the potential to adversely affect
sales of MP3 players, digital cameras, camcorders, handheld PCs, portable videogames, and GPS
devices, just as they have nearly eradicated the PDA market.1 For example, it was predicted that
global sales of GPS-enabled phones would reach 240 million in 200914, while 500 million of the
mobile phones shipped globally in 2007 were capable of playing digital music15. These
additional features found in mobile telephones can reduce the demand for other devices that
share those features and by extension reduce the demand of the BCs and wall adapters used to
power those applications.
For those mobile phones surveyed, DOE found all to use EPSs, while some have the
ability to be charged by USB or 12 V DC car chargers. EPS power output is low (2.5-5 W) and
nearly all are five volt output. Mobile phones use 3.7 V batteries with capacities that range from
3 to 5.6 watt hours.
Mobile phones have average lifetimes of 2 yearse. While the mobile phone itself is
designed to last longer than this, the mobile phone industry is driven by technological innovation
and trends, two factors that lead consumers to replace phones on a regular basis. Furthermore,
cell phone service contracts average two years in length; after this point, consumers are
frequently given the option of purchasing a replacement phone at a significant discount.
Therefore, in the past DOE considered mobile phone lifetimes to be two years and EPS lifetimes
to also be two years. However, the GSMA Universal Charging Solution, described below, will
increase the lifetime of the EPS.

GSMA Universal Charging Solution.


In early 2009, 21 mobile phone operators and manufacturers agreed to work together to
implement a universal battery charging standard for mobile telephones by 201216. Historically,
each mobile phone has been manufactured and sold with a unique EPS built specifically for that
phone and its internal battery. As a result, EPS unit shipments have mirrored mobile phone
shipments. This standard will eliminate the need for consumers to purchase a new EPS each time
a new mobile phone is purchased and, as a result, will reduce mobile phone EPS shipments.
e

DOE estimated a mobile phone lifetime of 2 years based on the standard length of time for mobile phone contracts.
Typically, after 2 years, consumers are able to replace mobile phones with newer models at significant discounts.

3-13

The Environmental Protection Agency (EPA) estimates that 125 million mobile phones
(and, by extension, nearly that many EPSs) are discarded annually in the United States17. The
universal charging agreement will use a micro USB interface and common output voltage so that
new chargers can work with multiple phones. This will result in a significant reduction in annual
EPS shipments after 2012, as the need to replace old EPSs when a new phone is purchased will
be eliminated. DOE forecasts that the number of battery chargers manufactured and sold in the
global market place (and, by extension, the United States) will be reduced by 50 percent.
The reduction in EPS shipments will be matched by a corresponding increase in product
lifetime for EPSs, which DOE estimates will be 4 years. The agreement also includes a no-load
mode power ceiling of 0.15 W.18 This no-load limit may reduce the energy consumption of
mobile phone EPSs. These potential impacts are discussed in section 3.2.3

Stationary Telephony
The stationary telephony subcategory includes cordless phones/answering machines,
VoIP adapters, and caller ID devices. All use wall adapters.
Cordless phones and answering devices are often packaged with multiple handsets (each
with a BC). A typical cordless phone set consists of a charging base with built-in answering
machine, a handset, and one or more satellite charging bases, each with its own handset. Each
charging base plugs into a wall outlet via a wall adapter to charge the batteries of the
corresponding handset. Hence, a cordless phone set will include between one and five wall
adapters and charging cradles. DOE estimates total EPS shipments for cordless phones and
answering devices was 39 million units in 2008.f Most cordless phone EPSs have output power
between 1.2 and 7.7 W and voltage between six and nine volts. DOE found cordless phone
batteries were either 2.4 or 3.6 volts and between 2.6 and 5.3 watt hours. Cordless
phone/answering machines have an average lifetime of 5.3 years.g
Voice over internet protocol, or VoIP, adapters are powered by EPSs, and 8.8 million
units in were shipped in 2008.1 VoIP adapters provide telephone service via an internet
connection. VoIP adapters typically have five or 12 V EPSs with power outputs of 10-14 W.
DOE assumes that stand-alone caller ID devices use EPSs similar to those used for cordless
phones. Only 460,000 units shipped in 2008. 1
The low and decreasing shipments of standalone caller ID devices can be explained by
convergence with stationary telephones, as many now incorporate caller ID technology.
One trend that negatively affects stationary telephony, including cordless
phones/answering machines and caller ID devices, is the decline in homes with landline
telephone service. With the increase in cellular telephone service indicated by increasing mobile
phone shipments, many households have opted out of landline telephone service. CEAs data
f

Based on data provided by the Consumer Electronics Association.

Based on the average values of three sources: PG&E et. al. (5 years) (PG&E et al., No. 20 at p. 10); Appliance

Magazine (5 years); and FY2005 Preliminary Priority-Setting Summary Report and Actions Proposed (6 years).

3-14

show that answering devices and cordless phone unit shipments have decreased by an average 17
percent annually since 2005. The exception in stationary telephony is the VoIP adapter
application, shipments of which have increased at a CAGR of 29 percent over four years. Strong
growth in mobile phone unit sales has offset declines in cordless phone unit sales, leading DOE
to conclude that, despite these fluctuations, the market for BCs and EPSs has remained largely
unchanged in this subcategoryh.
Table 3.2 Annual Telephone Unit Shipments (thousands) and CAGR
Application
2005
2006
2007
2008
VoIP
4,111
5,568
7,451
8,845
Adapters
Cordless
64,928
56,773
41,905
36,781
Phones/
Answering
Devices
Mobile
93,962
110,754
121,000
131,400
Phones
Total
163,001
173,095
170,356
177,026

CAGR (%)
29
-17

12
3

3.2.1.5 Household
This category encompasses a wide array of applications, ranging from water softeners to
digital cameras, and DOE estimates that 95 million of these applications that often use BCs or
EPSs shipped in 2008.

As cordless phone sales declined, manufacturers began to bundle multiple receivers and satellite bases in a
package. As a result, BC and wall adapter sales for this application are higher than cordless phone package sales.

3-15

Childcare

Baby Monitors
Breast Pumps
RC Cars (Hobby Grade)

Entertainment

RC Toys
Portable Video Game Systems
Video Game Consoles
Handheld Vacuums

Floorcare

Robotic Vacuums
Stick Vacuums
Electronic Pest Repellents

Household

Home Systems

Home Security Systems


Irrigation Timers
Water Softeners/Purifiers
Blenders

Kitchen
Appliances

Can Openers
Mixers
Camcorders

Photo/Video

Digital Cameras
Digital Picture Frames
Portable DVD Player
Air Mattress Pumps
Aquarium Accessories

Other Household

Indoor Fountains
Flashlights/Lanterns
Universal Battery Chargers

Household applications are grouped into seven subcategories based on intended use.
These subcategories differ significantly from one another in market and technology
characteristics and are, therefore, discussed individually below.

Photo/Video
The photo/video subcategory is comprised of consumer products primarily designed for
photography, video, and viewing pictures and movies. Applications included in this subcategory
are digital cameras, consumer camcorders, portable DVD players, and digital photo frames.
Shipments in the photo/video subcategory were 53.4 million units in 2008, with digital camera
shipments totaling 33 million.1 2008 shipments for the other applications in this category were
portable DVD players (8.4 million units12), digital picture frames (7.5 million units in 20081),
and camcorders (5.6 million units1). Digital cameras and camcorders have lifetimes averaging six

3-16

years and five years respectively, while DOE assumes that portable DVD players have lifetimes
of four years.
Growth in this category is driven primarily by growth in digital picture frame shipments,
which have increased from 1.5 million units in 2006 to 7.5 million units in 20081. Based on
inspection of 20 top-selling models, DOE estimates that close to 100 percent of digital picture
frames ship with EPSs. In the same time period, camcorder shipments increased from 5.3 million
units to 5.6 million units. 1 Despite slow growth, CEA predicts that low-cost camcorders with
solid state drives have the potential to boost future sales of this application. Furthermore, these
units tend to use BCs with USB (five volt) input. The popular Flip Video solid state camcorder,
made by Cisco, controls 17% of the camcorder market with its inexpensive line of camcorders
that use USB input to recharge their batteries. 19 DOE assumes 25% of camcorder shipments use
five volt input BCs. Shipments of digital cameras increased slightly from 32.9 million in 2006 to
33.2 million in 2008.1 CEA attributes this stagnated growth to a combination of market
saturation and an encroachment on sales due to convergence with other devices such as mobile
phones.
DOE estimates that 41 percent of digital cameras ship with BCs.20 Camcorder and digital
camera BC shipments have remained constant, while digital picture frames, which are relatively
new to the market, are rapidly gaining market share, resulting in a net increase of BC/EPS
shipments in the photo/video category.
Digital camera BCs typically provide output power at 4.2-8.4 V. They are used to
recharge batteries that typically have 3.2 to 11.1 watt hours of energy. The majority of
camcorders DOE surveyed used cradle chargers. DOE was unable to find data on aftermarket
digital camera wall adapters, but very few digital cameras use wall adapters. DOE identified one
model that used a wall adapter as part of the battery charging system. DOE is also aware of
digital cameras that have optional EPSs to power the digital camera without a battery. DOE
found a few camcorders that had the ability to draw power directly from the wall adapter with or
without an installed battery. In these cases, the wall adapter functions as an EPS. DOE also noted
a strong trend towards USB power for camcorders; the best selling model in 2008 was
rechargeable by this method19. DOE found EPSs for digital picture frames to range between 7.5
and 24 W of output power. Portable DVD players use wall adapters with output powers between
9 and 24 W to charge batteries with rated energy between 16 and 32 watt hours.
It is important to note that many mobile phones and PDAs now include digital camera
and digital video recording technologies. While DOE expects that this has had a negative effect
on the subcategorys shipments, this effect has most likely been small, a result of the relatively
poor quality of most mobile phone camera lenses and sensors. As mobile phones are equipped
with higher quality cameras with greater functionality, it is possible that the convergence will
increase. Portable DVD players may also face the pressures of convergence with other
applications. As streaming videos become more common and mobile devices (such as mobile
phones and portable music players) are able to store and play full-length digital movies, the
market for portable DVD players may decline. If these applications continue to converge, the
demand for multiple devices will be reduced, thus reducing the demand for BCs.

3-17

Another instance of convergence within this subcategory is the ability of many digital
cameras to record good-quality videos. Many compact, point and shoot digital cameras can
shoot videos, while a few recent entrants into the digital single-lens reflex (DSLR) camera
market can shoot high-definition videos. DOE expects that this convergence has reduced the
need for consumers to own separate digital cameras and camcorders and, as a result, has reduced
the need for the BCs and EPSs powering these devices.

Floor Care
This subcategory contains three applications: hand vacuums, stick vacuums, and robotic
vacuums. All three applications include models that utilize battery chargers and wall adapters
and include models with charging cradles.
In 2008, vacuum unit shipments were 4.2 million21 for rechargeable handheld models,
one million22 for robotic models (all are cordless, rechargeable models), and 4.2 million21 for all
stick vacuums. DOE estimates that 63 percent of stick vacuums (2.6 million) are rechargeable.
Thus, DOE estimates total BC shipments for floor care were 7.8 million units in 2008.
All of the handheld rechargeable vacuums surveyed are charged via charging cradles
coupled with wall adapters. BC and wall adapter specifications for handheld vacuums were not
readily available; however battery voltage ranged widely from 4.7 to 40 V, with most batteries
between 9.6 V and 20 V. DOE assumes typical hand vacuums have battery energy of 19 watt
hours.
Stick vacuums are designed for cleaning larger floor areas and have higher capacity
batteries than do handheld vacuums. Approximately 50-60 percent of the rechargeable units
surveyed were charged via a charging cradle, with the remainder utilizing wall adapters instead.
DOE expects that, like handheld vacuums, stick vacuums with charging cradles are designed for
the charger to be plugged into mains all the time. Charging times ranged from 3 to 24 hours,
while battery voltage ranged between 6 and 24 V.
The majority of popular robotic vacuums can be charged via a wall adapter and charging
base. The base is similar to a cradle in that the product spends the majority of its time plugged
in. Battery energy is higher for robotic models than for stick or handheld models, as the battery
must power the vacuum, sensors, and drive wheels. DOE inspected two additional robotic
products manufactured by iRobot, the largest manufacturer of robotic vacuums: a robotic floor
washer and a robotic gutter cleaner. Both use batteries of similar size and chemistry to the
companys floor vacuum. Charging time for robotic vacuums was between 3 and 15 hours.
Where data were available, DOE found BC output voltages of 17-22 V and energy between 36
and 43 watt hours.

Kitchen Appliances
Very few kitchen appliances use BCs or EPSs. DOE estimates that only about 400,000
shipments included BCs and just 20,000 had EPSs. Overall shipments of kitchen appliances are
significant, however. Appliance Magazine listed 2008 shipments for all electric blenders, can
openers, and mixers as 1.2, 5.7, and 5.8 million, respectively.21 The May 2009 issue of Appliance

3-18

Magazine shows that sales of electric handheld blenders, can openers, and mixers have been
steady since 200521. Conversations with the Association of Home Appliance Manufacturers
(AHAM) have supported DOEs assumption that rechargeable units make up small fractions of
the unit shipments for these applications.

Childcare
Baby monitors use BCs and wall adapters, while breast pumps use EPSs but not BCs.
Other applications considered elsewhere in the analysis are toys and entertainment devices. DOE
estimates that 2005 shipments of baby monitors were 1.5 million.23 i DOE assumes flat shipments
between 2005 and 2008 and that all units sold today are rechargeable. Most baby monitors use at
least two BCs and wall adapters, one for the nursery unit and one for the receiving unit. DOE
assumes that the average baby monitor is used intermittently over a period of 4 years. DOE does
not have shipment data for breast pumps, but estimates shipments 300,000 units annually. Breast
pumps are powered by EPSs.

Entertainment
The home entertainment subcategory includes toys, hobby products, and electronics
specifically designed for at-home recreational purposes. These include video game systems
(consoles and portable handheld systems), radio controlled toys, and hobby-grade radio
controlled vehicles. The ride-on toy application is included in the transport category due to
similarities between BCs for ride-on toys and BCs for other applications in that category. Other
applications not analyzed in this category include musical instruments, computers, and
recreational transport (such as motorized bicycles).
DOE estimates total unit shipments to be 34.5 million. The majority of units are
videogame systems. Shipments of the two most-popular handheld video game systems, the
Nintendo DS and Sony PSP, totaled 13.8 million units in 2008, and all use BCs and EPSs.24
DOE estimates lifetimes of 3 years based on the rate at which manufacturers typically develop
new systems. EPSs used with handheld game systems are 5 V and have output powers of 2.5 to
7.5 watts. They are powered by 3.7 V batteries with 1.8 to 4.4 watt hours of energy.
All of the top three video game consolesthe Nintendo Wii, Microsoft Xbox 360, and
Sony Playstation3use EPSs. Industry-wide shipments in 2009 were 18.4 million units.24
Shipments of the Nintendo Wii and Sony PlayStation 3, which are both powered by Class A
EPSs, totaled 13.5 million units; an additional 4.9 million non-Class A EPSs shipped with the
Microsoft Xbox 360, which uses a 198 watt non-Class A EPS that has multiple simultaneous
output voltages. According to a report published by the Natural Resources Defense Council, this
represents monthly sales growth of 8 percent over the previous 7 years25. EPS output power
ranges between 44.4 W and 48 W.
This category has experienced a form of convergence that may lead to an increase, rather
than a decrease, in EPS shipments. Most new videogame consoles have the ability to play DVD
and Blu-ray discs. Stationary DVD and Blu-ray players are not powered by EPSs. This additional
i

Estimate based on shipments of similar baby products.

3-19

functionality in video game consoles may cause some consumers who may otherwise have
purchased a stationary DVD or Blu-ray player to purchase a console instead, resulting in
accelerated console sales growth and more EPS shipments.
DOE lacks shipments data for radio-controlled (RC) toys and hobby-grade RC vehicles,
but assumes shipments of RC toys to be 7 million per year (similar to toy ride-on vehicles) and
250,000 hobby grade RC vehicles. DOE assumes that 30 percent of RC toys are rechargeable.
Hobby grade RC vehicles belong to a niche market. While some units use internal combustion
engines, DOE believes that half of the hobby grade RC vehicles on the market are shipped with
or powered by rechargeable battery packs. DOE estimates total shipments with BCs of both
categories to be 2.2 million units annually. Growth in BC shipments may result if there is an
increase in the ratio of rechargeable RC toys to those powered by primary batteries. Both types
of RC vehicles use cradle BCs and often use 7.2 or 9.6 volt batteries with energy between 7.7
and 10.8 watt hours.

Home Systems
Applications considered under the home systems category are designed to be
continually plugged into household power outlets. These applications operate as background
systems, adding comfort, security, or safety to homes. Home security systems, electronic pest
repellents, irrigation timers, and water softeners/purifiers are all included in this category. Many
use EPSs, and home security systems contain battery chargers. Water softeners/purifiers shipped
1.2 million units in 2008.26 Shipments data were not readily available for electronic pest
repellents and irrigation timers, but DOE estimates modest shipments of 800,000 units in 2008.
PG&E estimates lifetimes for emergency systems, which include home security systems,
to be 7.3 years. (PG&E et al., No. 20 at p. 10) Given the similarities among applications in this
category, DOE extended this estimate to the other applications in this category as well.
Given that 28 percent of homes contain security systems and assuming the lifetime provided by
PG&E, DOE estimates annual shipments of home security systems to be 4.2 million units. Home
security systems are comprised of various components that use combinations of rechargeable
batteries (with integrated chargers) and EPSs as power sources. While there are a few basic
configurations and do-it-yourself installation kits available, security systems are componentbased and highly customizable. As a result, the number of BCs and EPSs varies from system to
system. Security system EPSs tend to be AC-AC transformers, though some components are
powered by AC-DC converters. In the most basic home security systems, a simple, nonrechargeable battery-powered circuit is attached to a door, window, or other point of entry into a
home. Alarm control boxes that monitor these circuits are often powered by an EPS. An
integrated AC-DC converter that functions as a BC provides a continual source of power to a
backup battery so, in the event of a power outage, the security system remains functional.
Wireless systems can be completely battery powered (with optional after-market wall adapters)
and may include one or more BCs. As a simplification, DOE assumed one BC and EPS are
included with each security system.
Typical output powers for home security system EPSs (the majority of which are AC-AC
EPSs) are 16 or 24 V. DOE found that most home security system BCs convert AC power to DC
power in order to charge a security systems backup batteries. Many of these BCs convert AC

3-20

current to 6, 12, and/or 24 volt DC current to recharge the batteries.j Standard batteries are 12 V
sealed lead-acid with 1484 watt hours of energy. Some battery packs feature other battery
chemistries, 3.67.2 V output, and battery capacities of 1.313 Ah.
Water softeners surveyed used AC-AC EPSs rated at 24-volts and 9.6-18 watts nameplate
output. DOE found information on one irrigation timer that also had a 24 volt AC/AC EPS rated
at 18 watts.k Electronic pest repellants are typically powered by primary batteries or from mains.
Very few models use EPSs, and DOE expects shipments to be very low. BC/EPS data on
electronic pest repellants was not readily available.

Other Household
This subcategory is comprised of consumer applications designed for home use that do
not readily fall under the other subcategories. BC applications include air mattress pumps,
rechargeable flashlights, and universal battery chargers. Many of these applications also use wall
adapters, as do aquarium accessories (air and water pumps) and indoor fountain pumps.
PG&E estimated annual shipments of universal battery chargers to be 300,000 units and lifetimes
to be eight years. (PG&E et al., No. 20 at p. 11) BCC research predicts that sales of universal
battery chargers will continue to increase, driven primarily by an increase in rechargeable battery
sales27. BC data vary significantly based on the model of universal battery charger and the
batteries it is intended to charge. Simple models can accommodate only two or four AA
batteries, while others can charge most standard-size rechargeable primary batteries in various
combinations and quantities. A typical user may frequently charge four AA batteries; therefore
these BCs would typically charge batteries of ten watt hours.
PG&E estimates that 100,000 rechargeable flashlights are sold annually, a small fraction
of all the flashlights shipped annually. Rechargeable flashlights have expected lifetimes of ten
years. (PG&E et al., No. 20 at p. 11)
Aquarium and fountain pumps use AC/AC EPSs. DOE examined the ENERGY STAR
product list of AC/AC EPSs and found aquarium pump EPSs rated at 9 V and 1.8 watts of output
power. The American Pet Products Manufacturers Association estimates that 15 percent of
households have fish as pets.28 Assuming each fish owner has an aquarium with a pump and
given an estimated lifetime of five years, DOE estimates shipments of aquarium pumps to be 3.5
million units annually. DOE lacks a source for shipments data on indoor fountain pumps but
assumed annual shipments of one million units. DOE assumes indoor fountain pumps use EPSs
that are similar to aquarium pumps. DOE expects lifetimes for indoor fountains and air mattress
pumps to be approximately five yearsl, 29, 30.
3.2.1.6 Outdoor Appliances
j

Based on examination of retailer and distributor Web sites, including www.bassburglaralarms.com and

www.homesecuritystore.com.

k
The line of Toro ECXTR Sprinkler Timers uses a 24V, 18 watt AC/AC EPS.

l
Estimate for air mattress pumps and aquarium accessories are based on the lifetime estimate of 5 years for indoor

fountains. The EPS for indoor fountains primarily powers the fountains pump; the same holds true for the EPSs of
the other two applications. As a result, DOE assumes all three pump applications to have similar operational
lifetimes. Indoor fountain lifetimes estimates based on an average of the lifetimes quoted on retailer websites.

3-21

The market for battery powered, rechargeable outdoor appliances is small compared to its
gasoline-powered counterparts. DOE has identified three battery-powered outdoor appliance
applications: weed trimmers, hedge trimmers, and lawn mowers.

Weed Trimmers
Outdoor Appliances

Hedge Trimmers
Lawn Mowers

DOE lacks a source of shipments data for cordless weed trimmers and hedge trimmers
but estimates annual shipments to be 300,000 units. Although 300,000 electric lawn mowers
shipped in 2008, many of these units are corded. However, DOE believes the prevalence of
cordless mowers will increase. In fact, an examination of products available at major home
improvement stores showed the availability of 17 cordless lawn mowers compared to just 11
corded mowers.m Based on this model count, DOE assumes 61 percent of all outdoor appliances
use BCs. DOE estimates that, as technologies improve and battery capacity increases (allowing
the product to be used for longer periods of time), battery powered outdoor appliances could
experience significant growth.
Cordless lawn mowers require significant power and long discharge times. Most utilize
sealed lead-acid batteries and battery energy ranges from 240 to 840 watt hours, with a median
of 410 watt hours. Battery voltage was significantly higher than other outdoor appliances: most
electric lawn mowers use 24 V batteries, with some as high as 60 V. Most electric lawn mowers
have charging times of 12-24 hours.
Short discharge and long recharge times are limitations of cordless lawn mowers. Thus,
battery powered mowers are impractical for consumers with larger lawns. DOE expects the
market share for rechargeable units to increase as technologies improve discharge times and
charging rates.
By contrast batteries and battery chargers for cordless weed trimmers and cordless hedge
trimmers are much smaller. Both products have similar batteries with outputs between 12 V and
18 V and rated energy of 14 to 31 watt hours. Charging times for these products are relatively
short: between 1-3 hours.
3.2.1.7 Personal Care
The personal care products category includes three subcategories. The hair subcategory
includes beard and mustache trimmers, hair clippers, and electric shavers. The dental
m

Models of corded and cordless lawn mowers available on the websites of The Home Depot, Lowes, and Sears
were counted. Sites examined September 29, 2009.

3-22

subcategory includes rechargeable toothbrushes and rechargeable water jets (also known as oral
irrigators). These products tend to have shallow depths of discharge and smaller capacity
rechargeable batteries. DOE believes battery energy for most dental and hair products to be
approximately 1 watt hour. The medical subcategory includes blood pressure monitors, medical
nebulizers, portable oxygen concentrators, and sleep apnea machines. These applications use
non-class A EPSs.
Toothbrushes
Dental

Water Jets
Beard and Mustache Trimmers

Hair

Personal Care

Hair Clippers
Shavers
Blood Pressure Monitors
Medical Nebulizers

Medical

Portable Oxygen Concentrators


Sleep Apnea Machines

Unit shipments for applications in the hair subcategory were 24.1 million, divided
between trimmers (9.4 million), clippers (6.1 million), and shavers (8.7 million, 6.6 million of
which are rechargeablen, 31, 32). DOE believes the majority of trimmers use primary batteries,
while the majority of clippers are corded. DOE assumes that 25 percent of trimmers and clippers
use rechargeable batteries. DOE believes the markets for these products to be at or near
saturation. As a result, demand is expected to remain constant. Most rechargeable grooming
devices use a wall adapter, often in combination with a cradle. Although data is limited, DOE
observed that typical battery energy is between one and four watt hours. During its non-class A
EPS determination, DOE tested two personal care wall adapters. One was rated at 15 volts and
6.3 watts, while the other was rated at 3 volts and 0.3 watts of output power.
DOE currently lacks unit shipment data for oral care appliances, although research
published by the National Housewares Association indicates that 16 percent of oral care
appliances sold in the first quarter of 2009 were rechargeable. This is an increase from 15
percent in the first quarter of 2008.33 In 2000 ten million electric toothbrushes were sold.34 DOE
estimates the popularity of these toothbrushes has increased and assumes that about one percent
of the population purchases an electric toothbrush annually, leading to annual sales of
rechargeable toothbrushes at 4.9 million units. Currently rechargeable models in this subcategory
tend to be significantly more expensive than those powered by primary batteries. Rechargeable
oral care products are inductively charged and use cradles with wall adapters. Rechargeable
toothbrushes use inductively charged batteries. Their batteries are typically 1.2 volts and
approximately 0.8 watt hours.

Rechargeable shaver unit shipment estimates were calculated by multiplying total application shipments by the
International Housewares Associations estimate that 76 percent of shavers sold in Q2, 2007 were rechargeable.

3-23

In its notice of proposed determination for non-Class A EPSs DOE examined medical
devices designed for in-home use that employ EPSs. 74 FR 56928. Some of these applications,
including nebulizers, portable oxygen concentrators, and sleep apnea machines also use BCs.
Blood pressure monitors are used by those who must take frequent readings of their blood
pressure. Most digital units operate with primary batteries; however, some units are also sold
with an EPS or offer an optional EPS. DOE estimates typical nameplate output power to be
around 3 watts.
Nebulizers administer liquid medication as a mist that can be inhaled into the lungs. They
are commonly used to treat asthma and chronic obstructive pulmonary disease (COPD). The
EPSs that provide power to nebulizers tend to have nameplate output power in the range of 10 to
20 watts.
Sleep therapy devices include continuous positive airway pressure (CPAP), bi-level
positive airway pressure (biPAP), automatic positive airway pressure (autoPAP), and similar
machines used to treat obstructive sleep apnea. Some sleep therapy devices are battery powered,
some plug directly into mains, and others are powered by EPSs, which typically have nameplate
output power of approximately 28 to 50 watts. DOE found sleep apnea machines that employ 12
volt batteries with battery energy of 79.2 watt hours.
Portable oxygen concentrators absorb nitrogen from the air to provide oxygen to the user
at higher concentrations, eliminating the need for oxygen tanks. These devices typically use
EPSs ranging from 90 to 200 watts. Portable oxygen concentrators include batteries and are
typically sold with BCs for both at-home and in-vehicle use. Therefore, although DOE estimates
that approximately 58,000 portable oxygen concentrators were sold in 2008, the number of BCs
shipped with them is twice that. DOE found an example with a 195 watt hour battery.
3.2.1.8 Power Tools
The cordless power tool market is large, with 23.4 million units shipped in 2008.12 DOE
divides power tools into two categories: Do-It-Yourself (DIY) and professional tools.

DIY Power Tools (Integral)


Power Tools

DIY Power Tools (External)


Professional Power Tools

DIY tools are aimed at casual users and have batteries of less than 18 volts while
professional tools have batteries of 18 volts or more. Both types of tools are frequently purchased
by consumers.

3-24

DOE estimates that 50 percent of power tools shipped are DIY tools. These can be
divided into those with detachable batteries and those with integral batteries. DOE assumed that
the former account for 30 percent and the latter 20 percent of the total market. Based on data
provided by the Power Tool Institute, DOE estimated that 80 percent of DIY tools with
detachable batteries and 100 percent of DIY tools with integral batteries use wall adapters.
Professional power tools use detachable battery packs and the battery charging system does not
use a wall adapter. Based on manufacturer interviews and data from PG&E, DOE estimates
average power tool lifetime at 5.9 years for DIY tools and 5.4 years for professional tools.
(PG&E et al., No. 20 at p. 11)
According to forecasts from the Darnell Group, the market for cordless rechargeable
power tools will continue to grow at an average annual rate of 10.6 percent until 2013. This
growth is attributed to a falling cost for increasingly powerful and flexible tools. DOE believes
that short-term growth will be tempered by the slowdown in the construction and remodeling
industries.
Batteries for DIY tools are quite varied. Smaller tools, such as cordless screw drivers,
may have batteries in the 3.7 to 4.8 volt range, while larger tools such as drills have batteries
clustered around 7.2, 12, and 14.4 volts. Based on limited information, DOE estimates that
battery energy for DIY tool batteries fewer than 12 volts is typically less than 15 watt hours.o
DIY tools between 12 and 18 volts tend to have battery ratings around between 14 and 55 watt
hours.p Most professional power tools use 18 volt batteries. DOEs research found median
battery energy among professional tools to be 54 watt hours.q
3.2.1.9 Transport
The transport category includes an assortment of applications powered by BCs, including
toy ride-on vehicles, golf cars, mobility scooters, and others. While many electric vehicles are
included in this category, automobiles, as defined by the U.S. Department of Transportation, are
not. See chapter 2 for further explanation.
Electric Scooters
Electric Vehicles

Motorized Bicycles
Golf Carts
Toy Ride-On Vehicles

Transport
Mobility Devices
Large Battery
Chargers

Wheelchairs
Mobility Scooters
Marine/Automotive/RV
Chargers

DOE estimates that 8.6 million BCs for transport applications are shipped annually. Of
these, 7.1 million are for toy ride-on vehicles.35 While DOE assumes lifetimes for toy ride-on
o

Based on a sample of five DIY tools.


Based on a sample of five DIY tools.
q
Based on a sample of 13 professional tools.
p

3-25

vehicles to be about 4 years, other transport applications have lifetimes of approximately 10


years. (PG&E et al., No. 20 at p. 11) While 6-, 12-, and 24-volt batteries are common among
these applications, battery energy varies dramatically from 27 to over 9,000 watt-hours. The
output powers of the BCs and EPSs that power these batteries also vary considerably.

Electric Vehicles
The electric vehicles subcategory includes BCs for golf cars, chargers for marine and
recreational vehicle (RV) batteries, toy ride-on vehicles, motorized bicycles, and electric
scooters.
The golf car market includes a wide range of vehicle types, from standard golf carts to
heavy duty utility vehicles, which are designed to carry loads over difficult terrain. The
research firm International Market Solutions (IMS) calls this broad category of products, small
task-oriented vehicles and each vehicle type may be purchased by consumers. These vehicles
are steadily moving towards battery power. IMS estimates that the market share of electric small
task-oriented vehicles has increased from 56 percent in 2000 to 64 percent in 2006 and they
estimate the electric share will increase to 70 percent by 2012.36 The same golf car models are
often sold to both the golf course fleet market and to private consumers. Furthermore, many fleet
golf cars are later sold in the consumer market as used vehicles. IMS estimates that over 210,000
electric small task-oriented vehicles were shipped in 2008. Shipments are expected to grow at a
CAGR of 4.2 percent from 2006 to 2012. As mentioned above, DOE excludes golf cars
manufactured for on-road use. These vehicles are automobiles and fall outside DOEs scope.
Based on DOEs analysis of currently available products, most golf cars employ several 6 or 12
volt batteries and energy greater than 3,000 watt hours. Common golf car BCs have output
voltages of 36 or 48 volts.
DOE is also considering large universal battery chargers in its scope, such as those used
to charge batteries for marine trolling motors. DOE has found that these battery chargers are
functionally equivalent to those used to charge batteries for recreational vehicle (RV)
accessories, automotive and motorcycle starter batteries, and other applications. Marine and RV
applications use one or more 12 volt deep cycle batteries, depending on the requirements of the
accessories being operated. PG&E estimates 2009 shipments of large universal BCs to be
500,000 units and have lifetimes of ten years. (PG&E et al., No. 20 at p. 10) Large universal BCs
typically have output powers of 12 volts, although 24-volt universal BCs are not uncommon.
These BCs are used when more than one battery is being charged, as may be the case in some
marine and RV applications. Deep cycle marine batteries store approximately 830 watt hours of
energy.r
Toy ride-on vehicles make up the great majority of shipments battery charged transport
shipments. Toy Industry Association, Incorporated reported that 2008 retail sales in the U.S.
were $1.9 billion.37 Based on an estimated average retail price of $210, DOE estimated
shipments to be just over nine million units per year.s

Based on test unit.

DOE examined the retail prices of best-selling toy ride-on vehicles available from Wal-Mart, Toys R Us, and

Amazon.com. Web sites examined September 2009.

3-26

These vehicles have BCs with output voltages of 6, 12 and 24 volts and energy between
27 and 144 watt hours. Based on the recommended age levels for these products, DOE estimates
toy ride-on vehicles to have a service life of four years.t Since the market for these applications is
mature, and because the population of children age one to six is projected to grow at a compound
annual growth rate of only 0.65 percent during the analysis period, DOE does not expect
significant growth in this market.38
Electric scooters and motorized bicycles are the remaining applications DOE analyzed.
Based on recalls of toy scooters, DOE estimates that annual sales of electric scooters are at least
250,000 units per year.u DOE estimates annual shipments of electric bicycles to be 150,000 units
in 2008.39 The scooters and motorized bicycles DOE analyzed used batteries ranging from 24 to
48 volts. Batteries had rated energy between 108 and 456 watt hours. The Segway brand scooter
is unique in that it uses two 73.6 volt batteries (each at 427 watt hours of energy). DOE found
little information wall adapters for these applications.

Mobility Devices
Battery-powered wheelchairs and mobility scooters are common BC applications used by
individuals with mobility-limiting disabilities, obesity, arthritis and other medical conditions. In
early 2008, the market research firm Marketstrat, Inc. forecasted that 166,000 powered
wheelchairs and 192,000 mobility scooters shipped in 2008.40
Powered wheelchairs and mobility scooters use similar batteries and chargers. All of the
mobility devices examined by DOE were powered by two 12 volt batteries wired in series for a
total output of 24 volts. Battery energy for a single battery ranged from 144 to 900 watt hours
and with common devices employing pairs of either 144 watt hour batteries or batteries in the
range around 400 watt hours.
The market research firm Global Industry Analysts anticipates shipments of mobility
devices to grow by 10 percent CAGR between 2011 and 2015, primarily driven by an aging
population.41 Over the analysis period, the population aged 65 and older is expected to grow at
2.8 percent CAGRv, compared to 0.56 percent CAGR for the population under age 65.38
3.2.2

Shipments, Lifetimes, and Energy Performance


Awareness of annual product shipment trends and lifetimes is an important aspect of the
market assessment and the development of the standards rulemaking. For this rulemaking, DOE
relied on data from public sources, interested parties, industry reports, and its own estimates.

According to surveys of retailer websites, typical age categories are one to two years of age for low-powered
vehicles and two to six years of age for more powerful vehicles.
u
Based on recall data from the Consumer Product Safety Commission. Data show that recalls of individual models
account for significant shipments. For example see: U.S. Consumer Product Safety Commission. CPSC, Razor
USA Announce Recall of Electric Scooters. June 14, 2005. (Last accessed September 13, 2010.)
<http://www.cpsc.gov/CPSCPUB/PREREL/prhtml05/05193.html> The September 13, 2010 material from this
website is available in Docket # EERE-2008-BT-STD-0005. For more information, contact Ms. Brenda Edwards at
(202) 586-2945.

v
The annual growth rate of the over-65 population is estimated to be 3.36 percent in 2013. It will decline slightly

during the analysis period and is forecast to be 1.42 percent by 2032.

3-27

DOE used these data for three main purposes. First, DOE used the shipments data to select the
representative product classes and representative units for the engineering analysis. Generally,
DOE selected product classes and units to reflect the most common BC and EPS types used in
the United States today (see section 3.3 and chapter 5 of the preliminary analysis). Second, DOE
used the lifetime data in its lifecycle cost analysis (chapter 8) and to develop its inventory model
for the national impact analysis (see chapter 10 of the preliminary TSD).Third, the shipment data
and market trend information contributed to the shipments forecast used in the national impact
analysis (see chapter 9 of the preliminary analysis).
3.2.2.1 External Power Supply Shipments, Lifetimes and Energy Performance
DOE estimates that a total of 301 million EPSs shipped in 2008. Of these, 288.6 million
were Class A EPSs and 12.4 million were non-Class A EPSs. In 2005, an estimated 296 million
EPSs were shipped, which represents a compound annual growth rate of 0.6 percent. To estimate
shipments of EPSs for consumer products in 2005, DOE reanalyzed data from its Draft
Technical Report on BCs and EPSs.42 Table 3.3 shows the average lifetime and an estimate of
the number of units shipped in 2008 for each of four segments of product class A1, the three
other Class A EPS product classes, and the 11 non-Class A EPS product classes. No products
were identified in four of the non-Class A EPS product classes.

3-28

Table 3.3 External Power Supply Lifetimes and Shipments by Product Class
Average
Shipments
Output
Class ID
Lifetime
in 2008 (estimated)
Class A

AC-DC

Basic
Voltage

AC-AC

Low Voltage
Basic
Voltage
Low Voltage

A1 (0-10.25 W)
A1 (10.26-39 W)
A1 (40-90 W)
A1 (91-250 W)
A2

4.6
4.4
4.0
3.3
4.1

76,227,000
66,721,000
52,994,000
10,235,000
72,195,000

A3

7.3

7,994,000

A4

4.9

2,250,000

X1
X2
H1

5.0
5.0
4.4

3,782,000
4,901,000
3,000

M1

7.9

1,534,000

M2

12.0

25,000

M3

M4

B1

4.7

1,884,000

B2

4.4

275,000

B3

B4

Non-Class A
Multiple-Voltage
High Power

For Medical
Devices

AC
DC
AC
AC

For MotorAC
Operated BC
DC
Applications
and Detachable
AC
Batteries
AC

100 W
100 W
>250 W
Basic
Voltage
Low Voltage
Basic
Voltage
Low Voltage
Basic
Voltage
Low Voltage
Basic
Voltage
Low Voltage

The following 18 tables show EPS shipment estimates and lifetimes for the top
applications in each product class. Product class A1 is subdivided into four segments by
nameplate output power.

3-29

Table 3.4 EPS Product Class A1a: Top Applications, Shipments, and Lifetimes
2008 EPS Shipments
Average Lifetime
Top Applications by Shipments
(Years)
(Units)
1. Answering Machines

20,175,000

2. Cordless Phones

19,151,000

5.3

3. Mobile Phones

13,140,000

4. Portable Video Game Systems

10,884,000

7,660,000

Other

5,217,000

Total

76,227,000

4.6

5. In-Vehicle GPS

Other includes baby monitors, breast pumps, caller ID devices, clock radios, consumer two way radios, electronic
pest repellents, guitar effects pedals, keyboards, pre-amps, personal digital assistants, and wireless headphones.
Includes EPSs between 0 and 10.25 watts nameplate output power.

Table 3.5 EPS Product Class A1b: Top Applications, Shipments, and Lifetimes
2008 EPS Shipments
Average Lifetime
Top Applications by Shipments
(Years)
(Units)
1. LAN Equipment

27,581,000

2. VoIP Adapters

8,845,000

3. Digital Picture Frames

7,472,000

4. Portable DVD Players

7,140,000

5. MP3 Speaker Docks

7,012,000

Other

8,671,000

Total

66,721,000

4.4

Other includes camcorders, clock radios, computer speakers, external hard drives, image scanners, hedge
trimmers, portable printers, weed trimmers, and wireless speakers. Total may not match individual unit count due to
rounding. Includes EPSs between 10.25 and 39 watts nameplate output power.

3-30

Table 3.6 EPS Product Class A1c: Top Applications, Shipments, and Lifetimes
2008 EPS Shipments
Average Lifetime
Top Applications by Shipments
(Years)
(Units)
1. Notebook Computers

30,225,000

3.3

2. Video Game Consoles

13,512,000

3. Ink Jet Imaging Equipment

5,557,000

4. Netbook Computers

3,700,000

3.5

52,994,000

Total
Includes EPSs between 39 and 90 watts nameplate output power.

Table 3.7 EPS Product Class A1d: Top Applications, Shipments, and Lifetimes
2008 EPS Shipments
Average Lifetime
Top Applications by Shipments
(Years)
(Units)
1. Notebook Computers
2. LED Monitors
Total

10,075,000

3.3

160,000

10,235,000

3.3

Includes EPSs between 90 and 250 watts nameplate output power.

Table 3.8 EPS Product Class A2: Top Applications, Shipments, and Lifetimes
2008 EPS Shipments
Average Lifetime
Top Applications by Shipments
(Years)
(Units)
1. Mobile Phones

52,560,000

2. In-Vehicle GPS

7,660,000

3. MP3 Players

4,373,000

4. Portable Video Game Systems

2,893,000

5. Personal Digital Assistants (PDA)

1,582,000

Other

3,127,000

Total

72,195,000

4.1

Other includes Bluetooth headsets, digital cameras, consumer two way radios, e-book readers, electronic pest
repellents, handheld GPS units, handheld image scanners, rechargeable toothbrushes, and rechargeable water jets.

3-31

Table 3.9 EPS Product Class A3: Top Applications, Shipments, and Lifetimes
2008 EPS Shipments
Average Lifetime
Top Applications by Shipments
(Years)
(Units)
1. Home Security Systems

4,219,000

7.3

2. Aquarium Accessories

1,750,000

3. Water Softeners and Purifiers

1,150,000

11

4. Indoor Fountains

500,000

4.7

5. Irrigation Timers

375,000

10

7,994,000

7.3

Total

Table 3.10 EPS Product Class A4: Top Applications, Shipments, and Lifetimes
2008 EPS Shipments
Average Lifetime
Top Applications by Shipments
(Years)
(Units)
1. Aquarium Accessories
2. Indoor Fountains
Total

1,750,000

500,000

4.7

2,225,000

4.9

Table 3.11 EPS Product Class X1: Top Applications, Shipments, and Lifetimes
2008 EPS Shipments
Average Lifetime
Top Applications by Shipments
(Years)
(Units)
1. Ink Jet Imaging Equipment
Total

3,782,000

3,782,000

Table 3.12 EPS Product Class X2: Top Applications, Shipments, and Lifetimes
2008 EPS Shipments
Average Lifetime
Top Applications by Shipments
(Years)
(Units)
1. Video Game Consoles
Total

4,901,000

4,901,000

Table 3.13 EPS Product Class H1: Top Applications, Shipments, and Lifetimes
2008 EPS Shipments
Average Lifetime
Top Applications by Shipments
(Years)
(Units)
1. Amateur Radios
Total

3-32

3,000

10

3,000

10

Table 3.14 EPS Product Class M1: Top Applications, Shipments, and Lifetimes
2008 EPS Shipments
Average Lifetime
Top Applications by Shipments
(Units)
(Years)
1. Sleep Apnea Machines
1,000,000
8
2. Medical Nebulizers
450,000
7
Portable Oxygen Concentrators with
3.
50,000
11
lower output EPS (<100 watts
4. Blood Pressure Monitors
25,000
12
Portable Oxygen Concentrators with
5.
9,000
11
higher output EPS (>100 watts)
Total
1,534,000
7.9
Table 3.15 EPS Product Class M2: Top Applications, Shipments, and Lifetimes
2008 EPS Shipments
Average Lifetime
Top Applications by Shipments
(Years)
(Units)
1. Blood Pressure Monitors
Total

25,000

12

25,000

12

Table 3.16 EPS Product Class M3: Top Applications, Shipments, and Lifetimes
2008 EPS Shipments
Average Lifetime
Top Applications by Shipments
(Years)
(Units)
1. No Known Applications
Total

Table 3.17 EPS Product Class M4: Top Applications, Shipments, and Lifetimes
2008 EPS Shipments
Average Lifetime
Top Applications by Shipments
(Years)
(Units)
1. No Known Applications
Total

3-33

Table 3.18 EPS Product Class B1: Top Applications, Shipments, and Lifetimes
2008 EPS Shipments
Average Lifetime
Top Applications by Shipments
(Years)
(Units)
1. Toy Ride-On Vehicles

355,000

2. RC Toys

350,000

3. DIY Power Tools (External Battery)

281,000

5.9

4. DIY Power Tools (Integral Battery)

234,000

5.9

5. Handheld Vacuums

209,000

Other

457,000

Total

1,884,000

4.7

Other includes air mattress pumps, blenders, electric scooters, flashlights and lanterns, hair clippers, mixers,
motorized bicycles, robotic vacuums, shavers, stick vacuums, and universal battery chargers. Total does not match
individual unit count due to rounding.

Table 3.19 EPS Product Class B2: Top Applications, Shipments, and Lifetimes
2008 EPS Shipments
Average Lifetime
Top Applications by Shipments
(Years)
(Units)
1. Shavers

164,000

4.1

2. Beard and Moustache Trimmers

59,000

4.5

3. Hair Clippers

38,000

4. Can Openers

14,000

275,000

4.4

Total

Table 3.20 EPS Product Class B3: Top Applications, Shipments, and Lifetimes
2008 EPS Shipments
Average Lifetime
Top Applications by Shipments
(Years)
(Units)
1. No Known Applications

Total

Table 3.21 EPS Product Class B4: Top Applications, Shipments, and Lifetimes
2008 EPS Shipments
Average Lifetime
Top Applications by Shipments
(Years)
(Units)
1. No Known Applications
Total

3-34

Table 3.22 shows the distribution of EPS shipments by efficiency level in 2009. DOE
weighted tested units by their application shipments to derive efficiency distributions for product
class A1. The distributions for product classes A2, A3, and A4 are assumed to be equivalent to
those for product class A1. Although product classes M1, M2, B1, and B2 have EPSs of similar
construction to Class A EPSs, they are not subject to EISA standards. (DOE found no products in
M3, M4, B3, or B4.) DOE assumed that 10 percent of the market had efficiencies that fell below
EISA, while the remaining products had a similar distribution to Class A EPSs. The distributions
for product classes X1, X2, and H1 were taken from the determination analysis of non-Class A
EPSs. These distributions underlie the base case efficiency forecasts presented in chapter 9.
Table 3.22 Energy Performance of New External Power Supplies in 2009
Output

Class
ID

Percent of Market at Each CSL


CSL 0

CSL 1

CSL 2

CSL 3

CSL 4

CSL 5

A1

68

17

12

A2

68

17

12

A3

68

17

12

A4

68

17

12

X1
X2
H1

100
5
60

0
95
40

0
0
0

0
0
0

M1

10

58

17

12

M2

10

58

17

12

B1

10

58

17

12

B2

10

58

17

12

Class A
Basic
Voltage
Low
Voltage
Basic
Voltage
Low
Voltage

AC-DC

AC-AC
Non-Class A
Multiple-Voltage
High Power
For Medical
Devices

AC
DC

For MotorOperated BC
Applications AC
and
DC
Detachable
Batteries

<100 W
100 W
>250 W
Basic
Voltage
Low
Voltage
Basic
Voltage
Low
Voltage

3-35

DOEs Draft Technical Report contained estimates of EPS efficiency before federal
standards for Class A EPSs took effect in 2008. Table 3.23 shows the efficiency of units on the
market in 2007 from least efficient (Baseline Line Frequency EPS) to most efficient (Federal
Standard).
Table 3.23 Energy Performance of External Power Supplies in 2007
EPS Output Power
Efficiency Level
0 to < 4 W
4 to 60 W
(percent)
(percent)

> 60 W
(percent)

Baseline Line Frequency EPS

35

16

Baseline Switch Mode EPS

25

44

60

Federal Standard
(effective July 1, 2008

40

40

40

3.2.2.2 Battery Charger Shipments, Lifetimes and Energy Performance


DOE estimates that nearly 402 million BCs shipped in 2008. In 2005, 250 million BCs
were shipped, which represents a compound annual growth rate of 17 percent. To estimate
shipments of BCs for consumer products in 2005, DOE reanalyzed data from its Draft Technical
Report on BCs and EPSs.42
Table 3.24 shows, for each of the ten BC product classes DOE defined, the average lifetime and
an estimate of the number of units shipped in 2008.
Table 3.24 Battery Charger Lifetimes and Shipments by Product Class
Shipments in
Battery
Battery
Class
Average
2008
Energy
Voltage
ID
Lifetime
(estimated)
Inductive
1
5.0
5,354,000
Connection
2
3.2
206,221,000
<100 Wh <4 V
4<10 V
3
4.9
23,116,000
AC
10 V
4
4.6
69,758,000
DC
<20 V
5
5.5
2,333,000
100
3000 Wh 20 V
6
8.5
940,000
>3000
7
9.7
214,000
Wh
<9 V Input
8
3.3
71,825,000
DC-DC
9 V Input
9
5.1
15,519,000
AC Output
AC-AC
from
10
6.0
6,900,000
Battery

3-36

The following ten tables show BC shipment estimates and lifetimes for the top
applications in each product class.
Table 3.25 BC Product Class 1: Top Applications, Shipments, and Lifetimes
2008 BC Shipments
Average Lifetime
Top Applications by BC Shipments
(Years)
(Units)
1. Rechargeable Toothbrushes
2. Rechargeable Water Jets
Total

4,868,000

487,000

5,354,000

Total does not match individual unit count due to rounding.

Table 3.26 BC Product Class 2: Top Applications, Shipments, and Lifetimes


2008 BC Shipments
Average Lifetime
Top Applications by Shipments
(Years)
(Units)
1.
2.
3.
4.
5.

Mobile Phones
Answering Machines
Cordless Phones
Portable Video Game Systems
Digital Cameras
Other
Total

105,120,000
20,175,000
19,151,000
13,777,000
10,879,000
37,119,000
206,221,000

2
5
5.3
3
6
3.4

Other includes baby monitors, Bluetooth headsets, beard and moustache trimmers, can openers, e-book readers,
hair clippers, handheld GPS units, home security system components, MP3 players, MP3 speaker docks, PDAs, and
shavers.

Table 3.27 BC Product Class 3: Top Applications, Shipments, and Lifetimes


2008 BC Shipments
Average Lifetime
Top Applications by Shipments
(Units)
(Years)
1. Portable DVD Players
7,140,000
4
2. Camcorders
4,206,000
4.9
3. Toy Ride-On Vehicles
3,548,000
4
4. RC Toys
2,100,000
2
5. DIY Power Tools with external battery
1,753,000
5.9
Other
4,369,000
Total
23,116,000
4.5
Other includes air mattress pumps, blenders, DIY power tools with integral battery, handheld vacuums, hobby
grade RC cars, mixers, stick vacuums, universal battery chargers, and wireless speakers.

3-37

Table 3.28 BC Product Class 4: Top Applications, Shipments, and Lifetimes


2008 BC Shipments
Average Lifetime
Top Applications by Shipments
(Years)
(Units)
1.
2.
3.
4.
5.

Notebook Computers
Professional Power Tools
DIY Power Tools with external battery
Netbook Computers
Handheld Vacuums
Other
Total

40,300,000
11,688,000
5,259,000
3,700,000
2,797,000
6,014,000
69,758,000

3.3
5.4
5.9
3.5
6
4.2

Other includes flashlights and lanterns, medical nebulizers, portable printers, robotic vacuums, sleep apnea
machines, stick vacuums, toy ride-on vehicles, universal battery chargers, and weed and hedge trimmers.

Table 3.29 BC Product Class 5: Top Applications, Shipments, and Lifetimes


2008 BC Shipments
Average Lifetime
Top Applications by Shipments
(Units)
(Years)
1. Toy Ride-On Vehicles
1,774,000
4
2. Auto/Marine/RV Chargers
500,000
10
3. Portable Oxygen Concentrators
59,000
11
Total
2,333,000
5.5
Table 3.30 BC Product Class 6: Top Applications, Shipments, and Lifetimes
2008 BC Shipments
Average Lifetime
Top Applications by Shipments
(Units)
(Years)
1. Electric Scooters
250,000
10
2. Mobility Scooters
192,000
10
3. Lawn Mowers
182,000
6
4. Powered Wheelchairs
166,000
9.7
5. Motorized Bicycles
150,000
7
Total
940,000
8.7
Table 3.31 BC Product Class 7: Top Applications, Shipments, and Lifetimes
2008 BC Shipments
Average Lifetime
Top Applications by Shipments
(Units)
(Years)
Small Task Oriented Vehicles (Golf
1.
214,000
6.4
Cars)
Total
214,000
6.4

3-38

Table 3.32 BC Product Class 8: Top Applications, Shipments, and Lifetimes


2008 BC Shipments
Average Lifetime
Top Applications by Shipments
(Units)
(Years)
1. MP3 Players
39,358,000
4
2. Mobile Phones
26,280,000
2
3. Digital Cameras
2,720,000
6
4. Personal Digital Assistants (PDA)
1,779,000
4
5. Camcorders
1,402,000
4.9
Other
286,000
Total
71,825,000
3.4
Other includes Bluetooth headsets, e-book readers, and handheld GPS units. Total may not match individual unit
count due to rounding.

Table 3.33 BC Product Class 9: Top Applications, Shipments, and Lifetimes


2008 BC Shipments
Average Lifetime
Top Applications by Shipments
(Units)
(Years)
1. In Vehicle GPS
15,320,000
5
2. Medical Nebulizers
90,000
7
3. Portable Oxygen Concentrators
59,000
11
4. Flashlights and Lanterns
50,000
10
Total
15,519,000
5.1
Table 3.34 BC Product Class 10: Top Applications, Shipments, and Lifetimes
2008 BC Shipments
Average Lifetime
Top Applications by Shipments
(Units)
(Years)
1. Uninterruptible Power Supplies
6,900,000
6
Total
6,900,000
6
Table 3.35 shows the distribution of BC shipments by efficiency. DOE derived efficiency
distributions for product classes 1, 2, 4, 5, and 10 from an examination of the efficiency of tested
units. The efficiency distribution for product class 7 is based on manufacturer interviews
(discussed in chapter 5), while the distributions for product classes 3, 6, 8, and 9 are scaled from
those for the other product classes. These distributions underlie the base case efficiency forecasts
presented in chapter 9. DOE lacks historical data on BC efficiency.

3-39

Table 3.35 Energy Performance of New Battery Chargers in 2009


Percent of Market at Each CSL
Battery
Battery
Class
Energy
Voltage
ID
CSL 0
CSL 1
CSL 2
CSL 3
Inductive
1
Connection
<100
Wh
AC
DC
100
3000
Wh

AC-AC

3.2.3

12.5

12.5

<4 V

39.2

58

2.8

4<10 V

34

54

12

10 V

18.7

43

38.3

<20 V

43.4

16.3

40.3

20 V

50

20

30

50

20

30

<9 V Input

39.2

58

2.8

9 V Input

39.2

58

2.8

AC Output
from
Battery

10

88.9

11.1

>3000 Wh
DC-DC

75

Other Energy Efficiency Programs

There are many domestic and foreign energy efficiency programs designed to improve
the energy performance of BCs and EPSs. Those programs that might affect the United States
market are discussed below, first EPS and then BC programs. Information about these programs
informed DOEs base case efficiency forecasts, which it developed as part of the shipments
analysis (see chapter 9).
The first mandatory energy efficiency standards for EPSs were introduced in California
and Oregon in 2007. On December 19, 2007, the President signed into law the Energy
Independence and Security Act of 2007 (EISA 2007) (P.L. 110-140), which set a Federal
standard for Class A EPSs that took effect on July 1, 2008. Because the EPS market is global,
this standard led to improvements in the efficiency of EPSs sold worldwide. Furthermore, the
standard, while intended to regulate only Class A EPSs, is likely having a spillover effect on the
efficiency of BCs and non-Class A EPSs. The standard for Class A EPSs has increased the
demand for, and lowered the cost of, some of the more efficient components and has stimulated
the adoption of improved designs. Because some of the same techniques and components are
used to manufacture both Class A EPSs and other EPSs and BCs, DOE assumes that some of
these components and designs are being carried over into the design and manufacture of BCs and
non-Class A EPSs.
In the United States, manufacturers can use the ENERGY STAR label to differentiate
more-efficient EPSs from less efficient ones. Version 2.0 of the ENERGY STAR criteria for
EPSs took effect on November 1, 2008. As of September 15, 2009, there were already over

3-40

3,000 qualified models.43 EPA estimated that 47 percent of EPSs sold in the United States in
2008 met ENERGY STAR specifications.44
In April 2010 an EPS standard that is equivalent to the current Federal standard for Class
A EPSs took effect in the European Union. In 2011 a more stringent standard, equivalent to
version 2.0 of the ENERGY STAR criteria, will take effect. The Darnell Group estimates that the
E.U. will receive 33 percent of all EPS shipments in 2011, which is nearly equivalent to the
North American share of shipments. Given the size of the E.U. market, EPS standards there will
likely cause spillover effects, increasing the efficiency of EPSs sold in the United States.
A recent industry agreement for mobile phones known as the GSMA Universal
Charging Solution could drive down the energy consumption of EPSs used with these products.
The agreement incorporates a no-load (standby) power consumption requirement that is
stricter than both the current Federal standard and ENERGY STAR criteria.
ENERGY STAR is currently the only efficiency program for BCs in the United States.
Because the criteria, which took effect on January 1, 2006, do not cover active mode, they cannot
be directly compared to the CSLs in DOEs analysis. EPA estimates that 16 percent of BCs sold
in the United States met the ENERGY STAR criteria in 2008.44
In fact, there are no standards for BCs in effect anywhere that include requirements for
active mode, although many governing bodies have indicated their intent to harmonize around a
DOE BC standard (if the Department sets an energy conservation standard). Although not a
current regulation, the Pacific Gas and Electric Company developed a standard for BCs that
includes active mode, which it proposed to the California Energy Commission (CEC) for
adoption. Should California adopt this standard, it may impact the efficiency of BCs on the
market nationwide. Comparisons between the proposed PG&E BC standard and DOEs CSLs
are difficult to make because the PG&E standard is specified as a continuous function whereas
DOE defined separate CSLs for each product class. However, DOE found that in some product
classes, products at the best-in-market CSL would meet the PG&E standard and that in all
product classes, products at the max-tech CSL would meet the PG&E standard. When possible,
DOE intends to analyze BC units at an improved level above the best-in-market CSL.
DOE encourages interested parties to inform DOE of other upcoming or updated
programs that may impact the energy efficiency of BCs and EPSs sold in the United States.
Table 3.36 summarizes a number of voluntary and mandatory energy efficiency programs for
BCs and EPSs. For detailed information on these programs, please refer to appendix 3B.

3-41

Table 3.36 BC and EPS Efficiency Programs Worldwide


Country /
State

Program Name

Effective Date

Compliance

Coverage

US

EISA 2007

2008

Mandatory

EPS

ENERGY STAR

2006 (BC)
2008 (EPS)

Voluntary

BC, EPS

Tier II Standard for State


Regulated EPSs

2008

Mandatory

EPS

Battery Charger Standard

Under
Consideration

Mandatory

BC

Australia/
New Zealand

Minimum Energy
Performance Standards

2008 (AU)
2009 (NZ)

Mandatory
(Mark III)
Voluntary
(Mark IV, V)

EPS

Canada

Canadian Standards
Association

2010
(projected)

Mandatory

EPS

China

China National Institute of


Standardization (CNIS)

~2010

Mandatory

EPS

China Standard
Certification Center (CSC)

2005

Voluntary

EPS

Energy using Products


(EuP)

2010 version 1
2011 version 2

Mandatory

EPS

EU Code of Conduct

2009

Voluntary

EPS

Group for Energy Efficient


Appliances

2007

Voluntary

BC, EPS

Manufacturers
Agreement

GSMA Universal
Charging Solution (mobile
phones)

2012

Voluntary

EPS

Israel

SI 4665.2

2007

Voluntary

EPS

California

European
Union

3-42

Country /
State

Program Name

Effective Date

Compliance

Coverage

Korea

Minimum Energy
Performance Standards

2009

Mandatory

EPS

e-Standby

2007

Voluntary

BC, EPS

3.2.4

Production and Distribution

DOEs BC and EPS distribution models were created based on information gathered
through market research, conversations with industry experts, and stakeholder feedback. The
Association of Home Appliance Manufacturers and others commented that, despite several ways
to market, most BCs and EPSs follow a similar distribution path. The most common path to
market, as identified by DOE, is depicted by the gray arrows in Figure 3.4, while alternative
paths are depicted by the white arrows. The distribution channels DOE identified are discussed
below.
BC and EPS distribution begins with component manufacturers, who produce the
circuitry, circuitry components, wiring, housing, and other materials needed to manufacture BCs
and EPSs. DOE learned that demand for specific components can drive their prices down. This is
sometimes the case for components used to make a BC or EPS more efficient. Given greater
demand for efficient components, due to an efficiency standard, for example, component
manufacturers increase production and the increased scale causes prices to fall.
Components are often sold directly to BC/EPS manufacturers, who produce a finished
BC or EPS, often for a specific end-use product manufacturer. Although less common, some
BCs or EPSs may be manufactured directly by the end-use product manufacturer (OEM). DOE
does not have data on the total size of the BC industry, but the Darnell Group estimated the size
of the EPS industry. It estimated that in 2005, over 300 manufacturers worldwide made EPSs.
Most of these manufacturers are located abroad. In the aggregate, their revenues totaled
$5 billion in 2005 and $6.7 billion in 2008.45,12 It should be noted that many of these
manufacturers also produce other products, including BCs and internal power supplies, so it is
difficult to get an exact value of EPS market size.

3-43

Component Manufacturer

BC/EPS Manufacturer

BC/EPS Distributor

End-Use Product Manufacturer (OEM)

Consumer Product Retailer/Distributor

Taxes

Consumer
*Note that widths of arrows are not drawn to scale and are not meant to be an exact indication of a distribution
paths relative prominence.

Figure 3.4 Paths of Distribution for Battery Chargers and External Power Supplies
BCs and EPSs are then typically purchased by an end-use product manufacturer,
henceforth known as the original equipment manufacturer, or OEM, as an input to an end-use
consumer product. The BC and/or EPS is typically packaged with a consumer product, or
especially in the case of some BCs, integrated into the consumer product.
Retailers typically purchase BCs and EPSs from OEMs and sell the products to
consumers, though DOE has identified a number of instances where the manufacturing and retail
operations for a product are owned and managed by one company. An example is Apple, which
manufactures its own consumer electronics for sale in its own Apple-branded retail stores.
In addition to the standard distribution chain described above, market research and stakeholder
comment revealed additional BC and EPS distribution channels. These are discussed below.
DOE found that many OEMs with low production volumes opt to purchase BCs or EPSs from
distributors because they provide easy access to a wide array of components. Because sourcing
BCs and EPSs through a distributor may be more costly, most OEMs with larger production
volumes eliminate this step by working directly with component and BC/EPS manufacturers.
DOE also notes that while most consumer products are manufactured in an OEM-owned factory,
there is a trend towards the use of electronics manufacturing services (EMSs). OEMs can take
advantage of greater economies of scale in source materials and components by contracting out

3-44

the manufacture of specific consumer applications to an EMS. EMSs achieve these economies of
scale by producing similar products for several OEMs.
3.2.5

Small Businesses

During this rulemaking process, DOE is considering the possible impacts to small
businesses that may be imposed by increased energy conservation standards for battery chargers
and external power supplies. The Small Business Administration (SBA) determines appropriate
guidance as to what is considered a small business for all industries described under the North
American Industry Classification System (NAICS)w. BC and EPS manufacturers fall under
NAICS code 335999 (All Other Miscellaneous Electrical Equipment and Component
Manufacturing). Based on this code, SBA defines a BC or an EPS manufacturer to be a small
business if it employs no more than 500 employees. Throughout its analysis of the BC/EPS
market space, DOE will attempt to gauge the prevalence of small businesses within these
industries.
3.2.6

Manufacturers and Market Shares


The Darnell Group estimated revenues of the top EPS manufacturers in their 2005 report
on the EPS market. They noted that the great majority of EPS manufacturing takes place in
China and Taiwan. Furthermore, many of the largest manufacturers are also based in Asia.
Figure 3.5 shows the top EPS manufacturers worldwide in 2005.45 Because of the global reach of
the industry and the reliance of major manufacturers on producing for high volume applications,
Darnell notes that there are very few differences in regional market shares. In 2005, there were
over 300 manufacturers producing EPSs. DOE learned that the industry has seen consolidation,
but it found that the manufacturers shown in Figure 3.5 remain independent of one another. None
of the top manufacturers listed in Figure 3.5 are headquartered in the United States. DOE
identified SL Power Electronics, a subsidiary of SL Industries, Inc., as an EPS manufacturer
based in the United States. SL Power develops, manufactures, and markets products under the
brand names CONDOR and AULT. DOE also recognizes that some EPSs for niche
applications may be manufactured in the United States.

For a more detailed description of SBAs small business definitions, see


http://www.sba.gov/idc/groups/public/documents/sba_homepage/sba_010224.pdf

3-45

Percent of 2005 Revenue


Delta Electronics 9.4%
Lite-on Technology 7.2%
Astec 5.8%
FRIWO Group 5.1%
Phihong 5.0%

$5
Billion

Salcomp 3.6%
Leader Electronics 2.8%
AcBel Polytech 2.1%
FSP Group 1.9%
Dee Van Enterprises 1.8%
Others 55.3%

Figure 3.5 Top EPS Manufacturers in 2005

Battery charger manufacturing is split between companies that produce BCs for OEMs
and OEMs that produce BCs in house. DOE currently lacks market share information for BC
manufacturers but it gathered some data from its manufacturing interviews (discussed in chapter
5). DOE learned that in most cases low-energy BCs are not produced by the OEM of the end-use
product; rather, they are purchased from an original device manufacturer (ODM), supplier, or
vendor typically based in Asia. Conversely, medium and high energy products BCs, such as
those for wheelchairs and golf cars, are typically manufactured in the U.S. For example, Lester
Electrical manufactures BCs for both wheelchairs and golf cars and has U.S.-based
manufacturing. Xantrax Technology Inc. is based in Canada, but has facilities in the U.S. They
produce BCs for marine applications. Many power tool OEMs also have some U.S.-based
manufacturing. Companies include Black & Decker, TTI (maker of Milwaukee, Ryobi, and
Hoover brand products), and the Robert Bosch Tool Corporation (Bosch, Skil).
3.2.7

Trade Associations and Other Interested Parties

DOE has identified a number of organizations that may have an interest in this
rulemaking. Energy efficiency advocacy organizations with a demonstrated interest in DOEs
rulemakings on battery chargers and external power supplies include
Appliance Standards Awareness Project,
American Council for an Energy-Efficiency Economy,
Earthjustice,
Natural Resources Defense Council,
Pacific Gas and Electric Company, and

3-46

Southern California Edison.


There are a substantial number of trade associations with member companies that
manufacture or sell BCs, EPSs, or the consumer products they power. DOE has identified 39
such trade associations, listed in Table 3.37 along with the products that DOE believes each
association has an interest in.
Table 3.37 Trade Associations
Association Name

Products and Applications

Alarm Industry Communications Committee


Amateur Radio Relay League (ARRL)
American Association of Cleaning Equipment
Manufacturers
Association of Home Appliance Manufacturers
(AHAM)
Battery Council International
Cellular Telecommunications Industry Association
(CTIA)
Computer and Communications Industry Association
Consumer Electronics Association (CEA)
Craft and Hobby Association
Electric Drive Transportation Association
Electronic Components Association
Hobby Manufacturers Association
Information Technology Industry Council (ITIC)
International Disk Drive Equipment and Materials
Association
International Housewares Association
International Music Products Association
International Recording Media Association
Irrigation Association
Juvenile Products Manufacturers Association
Motorcycle Industry Council
Multifunction Products Association
National Association of Manufacturers (NAM)
National Bicycle Dealers Association
National Burglar & Fire Alarm Association
National Electrical Manufacturers Association
(NEMA)
National Gardening Association
National Marine Manufacturers Association
3-47

Home Security Systems


Amateur Radios
Floor Care Appliances
Home Appliances
Batteries
Cell Phones
Computers and Peripherals
Consumer Electronics
RC Cars (Hobby Grade)
Electric Vehicles
Battery Chargers, External Power
Supplies
RC Cars (Hobby Grade)
Computers and Peripherals
External Media Drives, External Hard
Drives
Kitchen Appliances, Floor Care,
Personal Care
Keyboards, Guitar Effects Pedals,
Electric Music Instruments
International Recording Media
Association
Irrigation Timers
prenatal to preschool electronics
Electric Scooters
Multifunction Devices (MVD's)
All
Electric Bicycles
Home Security Systems
All
Outdoor Appliances
Marine Electronics

Association Name

Products and Applications

National Pest Management Association


National Retail Federation
Photo marketing Association
Portable Computer and Communications Association

Electronic Pest Repellents


All
Photo marketing Association
Notebooks, Netbooks, Handheld
Computers, Mobile Phones, Bluetooth
Batteries
Batteries, Power Supplies
DIY Power Tools, Professional Power
Tools
Home Security Systems
ALL
Telephony
Electric Wheelchairs
RC Toys, Toy Ride-On Vehicles

Portable Rechargeable Battery Association (PRBA)


Power Sources Manufacturers Association (PSMA)
Power Tool Institute (PTI)
Security Industry Association
TechAmerica
Telecommunications Industry Association
The National Mobility Equipment Dealers Association
Toy Industry Association

3-48

3.3

PRODUCT CLASSES
When necessary, DOE divides covered products into classes by the type of energy used,
the capacity of the product, and any other performance-related feature that justifies different
standard levels, such as features affecting consumer utility. (42 U.S.C. 6295(q)) DOE then
conducts its analysis and considers establishing or amending standards to provide separate
standard levels for each product class. This section is a continuation of the discussion in chapter
2 where DOE summarized comments submitted by interested parties relating to product classes.
3.3.1

EPS Product Classes

The following sections summarize all of the factors that DOE considered as potential
determinants for EPS product classes. When discussing EPS efficiency DOE refers to efficiency
as the matched pairing of active mode average efficiency and no-load mode power consumption.
3.3.1.1 Nameplate Output Power
Nameplate output power is a measure of the maximum power that an EPS can deliver,
which directly impacts capacity and efficiency. EPSs with greater nameplate output power offer
the consumer greater capacity and tend to have higher active mode average efficiency. EPSs with
lower nameplate output power tend to have lower no-load power consumption.
EPS active mode average efficiency reflects the power consumption (loss) within an EPS,
which comes from two sources: conversion losses and overhead losses. Conversion losses are
proportional to the power that the EPS outputs whereas overhead losses are essentially fixed
losses that do not increase significantly once output power is greater than 50 watts. Therefore,
EPSs with higher output powers have proportionally lower overhead losses and are more
efficient, when compared to EPSs with lower output power. In contrast to average efficiency,
EPS no-load power consumption improves (is less) for EPSs with nameplate output power less
than 50 watts because those EPSs have lower overhead requirements and can therefore shut
down more fully when not providing output power. Because of these factors, both the EISA
standard and Energy Star 2.0 specification determine a minimum efficiency level as a continuous
function of nameplate output power, as shown by the average efficiency levels in Figure 3.6 and
the no-load power levels in Figure 3.7. DOE acknowledges that nameplate output power
significantly affects utility and efficiency and therefore different nameplate output powers may
warrant unique product classes. However, rather than create distinct product classes by
nameplate output power, DOE proposes to follow the precedent set by EISA and Energy Star 2.0
such that DOE intends to set an efficiency standard level as a continuous function of nameplate
output power.

3-49

Minimum AVerage EFficiency in ACtive Mode [%]

100%
90%
80%
70%
60%

EISA

50%

Energy Star 2.0 Regular Voltage

40%

Energy Star 2.0 Low Voltage

30%

20%
10%
0%
0

50
Nameplate Output Power [watts]

Figure 3.6 The EISA Standard and Energy Star 2.0 Specification for Average Efficiencyx

Maximum Power Consumption in No-Load Mode [watts]

0.6

0.5

0.4

EISA

0.3

Energy Star 2.0 AC/DC

Energy Star 2.0 AC/AC


0.2

0.1

0
0

50
Nameplate Output Power [watts]

Figure 3.7 The EISA Standard and Energy Star 2.0 Specification for No-Load

Power Consumption

Energy Star 2.0 describes the two specification levels as standard and low voltage. Because DOE uses
standard as a term of art, the Energy Star 2.0 standard level is referred to as the Energy Star 2.0 regular level
throughout this document.

3-50

3.3.1.2 Nameplate Output Voltage


Nameplate output voltage affects utility because the main function of an EPS is to
provide an application with power at a certain voltage rather than 115 volts provided by mains.
The specific nameplate output voltage is determined by the characteristics of the application. For
instance, certain applications such as modems and computer monitors use have digital circuitry
that requires specific power at a specific voltage, such as 12V. For these applications the EPS
provides power at the necessary voltage. Therefore, output voltage offers consumers a distinct
utility that affects efficiency, which is one of the factors highlighted for special consideration
under 42 U.S.C. 6295(q)(1).
EPSs with higher output voltage will tend to be more efficient. This arises because of the
relationship between power, voltage, and current: power (P) = current (I) voltage (V). For an
EPS with a given output power, the voltage decreases as current increases. This is important
because many of the losses in a BC or an EPS are functions of current. For instance, the resistive
loss through a wire is I2 R and the power consumption of a diode is I Vdiode. Table 3.38
illustrates this phenomenon: EPS As output voltage is half that of EPS B, but EPS As I2 R
losses four times as high. Although the example is for a 20-watt EPS, it is applicable to all EPSs
because they all have I2 R losses.
Table 3.38 Comparison of I2 R Losses for Two 20-watt EPSs
Nameplate Output Voltage Nameplate Output Current
[volts]
[amps]
EPS A 10
2
EPS B 20
1

I2 R Losses
[watts]
4R
1R

Energy Star 2.0 acknowledged the relationship between voltage and efficiency by setting less
stringent active mode average efficiency criteria for EPSs with low voltage and high current
output (Figure 3.6). Energy Star 2.0 defined low voltage models as EPSs with nameplate
output voltage less than six volts and nameplate output current greater than or equal to 550
milliamps. Figure 3-8 shows the distribution of EPSs qualifying for Energy Star 2.0 as of 2009;
many low-voltage EPSs would not have qualified at the regular level. DOE proposes to have
different product classes for EPSs with regular voltage output and low voltage output. DOE
proposes to use the Energy Star criteria to define low voltage EPSs because the criteria
adequately capture many low voltage EPSs in the market and to follow precedent.

3-51

EnergyStar- Most Efficient AC-DC EPSs - Sept 2009


100%

90%

Standard EPSs

Efficiency [%]

80%

Low Voltage EPSs


EnergyStar 2.0
Standard
EnergyStar 2.0 LowVoltage
EISA Level

70%

60%

50%

40%
0

20

40

60

80

100

Power Out [Watts]

Figure 3-8 EPSs Qualifying for Energy Star 2.0 as of September 2009.
3.3.1.3 Type of Power Conversion (AC/AC versus AC/DC)
All EPSs receive input power from mains in the form of alternating current (AC) and provide
output power in the form of either AC power or direct current (DC) power. This is another key
functionality of an EPS, along with providing power at a specific voltage. Again, the type of
power conversion the EPS provides is governed by its application. Applications such as mobile
phones and laptops require DC power to match the type of power provided by the battery.
Applications that tend to use EPSs that provide AC power, such as cordless phones, often have
conversion circuitry within the application for AC-DC conversion.
The type of power conversion is indicative of an EPSs internal circuitry, hence its capability to
conserve energy. EPSs that provide AC output power typically consist of just a transformer. The
no-load power losses of those EPSs are dominated by transformer core losses. Alternatively,
EPSs that provide DC power output typically contain a transformer as well as overhead circuitry
that control the flow of power through the EPS . Overhead circuitry provides EPSs with DCoutput power the ability to reduce power consumption in no-load mode whereas EPSs with ACoutput power do not typically contain overhead circuitry. Energy Star 2.0 acknowledges this
relationship by setting a less stringent no-load mode power consumption criterion for EPSs with
AC output power (Figure 3.7). DOE proposes to have different product classes for EPSs with AC
output power and DC output power.
3.3.1.4 Use with Medical Equipment
Another condition by which EPCA defines a Class A EPS is that it does not include any device
that requires Federal Food and Drug Administration listing and approval as a medical device in
3-52

accordance with section 513 of the Federal Food, Drug, and Cosmetic Act (21 U.S.C. 360c).
(42 U.S.C. 6291(36)(C)(ii)(I)) Thus, all EPSs used with medical devices must meet the special
requirements of UL 60601 (Underwriters Laboratories standard for power supplies for medical
devices) such that they are approved by the Federal Food and Drug Administration (FDA). Use
with medical devices is a utility that is unique to medical EPSs. Medical EPSs are exempted
from EISA 2007 efficiency standards; therefore DOE expects that there may be medical EPSs in
the market with efficiencies lower than required by EISA 2007 for Class A EPSs.
3.3.1.5 Use with Battery Chargers of Motorized Applications and Detachable Batteries
A further condition by which EPCA defines a Class A EPS is that it does not power the charger
of a detachable battery pack or charges the battery of a product that is fully or primarily motor
operated. (42 U.S.C. 6291(36)(C)(ii)(II)) Thus, per EPCA, DOE considers EPSs for battery
chargers of motorized applications and detachable batteries" (MADB EPSs) to have a unique
utility from Class A EPSs. Products that are fully or primarily motor operated include portable
rechargeable household appliances such as handheld vacuums, personal care products such as
shavers, and power tools. MADB EPSs are exempted from EISA 2007 efficiency standards;
therefore DOE expects that there may be MADB EPSs in the market with efficiencies lower than
required by EISA 2007 for Class A EPSs.
3.3.1.6 EPS Product Classes Under Consideration
For all the reasons discussed above, DOE is considering the EPS product classes listed in Table
3.39, Table 3.40, Table 3.41, Table 3.42, and Table 3.43. Under these product classes, an EPSs
product class is determined by the type of power conversion it provides as well as its type of
output voltage. DOE proposes using the same criteria as Energy Star 2.0 to distinguish low
voltage EPSs from regular voltage EPSs, as discussed in section 3.3.1.2. Within each product
class, the standard level will vary with nameplate output power and output cord length. Chapter 5
of this preliminary TSD provides detail on the selection of candidate standard levels.
Table 3.39 Proposed Class A EPS Product Classes.
Regular Voltage Output Low Voltage Output *
AC-DC Conversion A1
A2
AC-AC Conversion A3
A4
* Low voltage output EPSs have nameplate output voltage less than six volts and nameplate output current greater
than or equal to 550 milliamps. All other EPSs are regular voltage output.

Table 3.40 Proposed Multiple Voltage EPS Product Classes.


#
Nameplate Output Power < 100 watts
X1
Nameplate Output Power 100 watts
X2
Table 3.41 Proposed High Power EPS Product Classes.
#
Nameplate Output Power > 250 watts
H1

3-53

Table 3.42 Proposed Medical EPS Product Classes.


Basic Voltage Output
Low Voltage Output *
AC-DC Conversion
M1
M2
AC-AC Conversion
M3
M4
* Low voltage output EPSs have nameplate output voltage less than six volts and nameplate output current greater
than or equal to 550 milliamps. All other EPSs are basic voltage output.

Table 3.43 Proposed MADB EPS Product Classes.


Basic Voltage Output
Low Voltage Output *
AC-DC Conversion
B1
B2
AC-AC Conversion
B3
B4
* Low voltage output EPSs have nameplate output voltage less than six volts and nameplate output current greater
than or equal to 550 milliamps. All other EPSs are basic voltage output.

3.3.2

BC Product Classes

In this section, DOE will present the BC characteristics it considered for use in setting
BC product classes, the impacts of power converter topology on BC design, and finally, the
resultant product classes DOE used in the preliminary analysis.
3.3.2.1 BC Product Class Criteria Considered for the Preliminary Analysis
In chapter 2, DOE summarized the comments it received from interested parties
regarding product classes. DOE then evaluated the criteria for setting product classes suggested
in the comments against the requirements of 42 U.S.C. 6295(q), resulting in the following BC
characteristics for further evaluation as criteria for setting product classes:
output power;
battery voltage;
battery capacity;
battery energy; and
inductive charging.
The above factors were combined with these additional factors that DOE took into account based
on its review of BCs in the market:
input voltage type (line AC or low-voltage DC); and
AC output.
Output power, battery voltage, battery capacity, and battery energy apply to all BCs, and

are related through the following equations, which were presented previously in chapter 2.

P_max >~ E_batt r,

E_batt = C_batt V_batt

3-54

Where:
P_max is the maximum output power, in watts
E_batt is battery energy, in watt-hours
r is the charge rate, in units of C or 1/hour
C_batt is the battery capacity, in ampere-hours
V_batt is the battery voltage, in volts
As can be seen in the above equations, the four BC characteristics are related. BC output
power is primarily affected by the charge rate and the battery energy, which in turn, is the
product of battery capacity and voltage.
Of these BC characteristics, DOE is using battery voltage and battery energy as a primary
means of dividing product classes for the preliminary analysis. In addition, DOE is also using
input and output characteristics, such as inductive charging, input voltage type (line AC or lowvoltage DC), and AC output to divide BCs into further product classes.
As mentioned earlier in chapter 2, battery voltage greatly affects consumer utility,
because the electronics of a portable consumer product are designed to require a particular
battery voltage. Whereas change in battery capacity would impact the runtime of a batteryoperated product, a change in battery voltage may stop it from running altogether. Furthermore,
BCs charging lower-voltage batteries tend to be less efficient, and could be disproportionately
affected by an equally stringent standard level across all voltages. Therefore, DOE is using
battery voltage and not battery capacity as a characteristic for setting product classes in the
preliminary analysis.
Whereas battery voltage specifies which consumer product applications can be used with
a particular battery (and its corresponding BC), battery energy describes the total amount of
work that the battery can perform, regardless of the application. Battery energy is therefore also a
measure of utility. Furthermore, because a BC must provide enough output power to replenish
the energy discharged during use, the capacity and physical size of the BC depend on the battery
energy.y By using battery energy as a proxy for output power, DOE is using one criterion for
classifying BCs instead of two, simplifying the potential BC energy conservation standards while
sufficiently accounting for any differences in BC capacity or utility in the standards analysis.
Finally, DOE is also using the presence of inductive charging, DC input voltage, and AC
output from the battery as additional characteristics for setting product classes. Since only a
small subset of BC products have inductive charging, DC input voltage, or AC output from the
battery, product classes based on these criteria typically fall within one of the larger classes
based on battery voltage and energy mentioned above.

The minimum output power is a product of battery energy and charge rate. However, while charge rates rarely fall
outside the range of 1 C to 10 C, the battery energy of consumer BCs can span over 5 orders of magnitude from
1 watt-hour to over 10,000 watt hours. Therefore, the output power is more dependent on battery energy.

3-55

Inductive charging was described previously in chapter 2 as a utility-related characteristic


designed to promote cleanliness and guarantee uninterrupted operation of the BC in a wet
environment.
Input voltage type was not identified by interested parties as a potential characteristic for
setting product classes. Nonetheless, while conducting an analysis of the market for BCs, DOE
identified numerous BCs that do not include a wall adapter, connecting instead to a personal
computers USB port or a cars cigarette lighter receptacle. Because input voltage can have a
differential impact on BC efficiency and, furthermore, because input voltage determines where
the BC can be used, impacting utility, DOE is using this characteristic as a criterion for
developing further product classes beyond the ones specified above. BCs differentiated on the
basis of the above criteria have been further divided based on input voltage in the preliminary
analysis to account for the efficiency losses associated with the AC-DC conversion process.

3.3.2.2 Impacts of Topology on Product Class Selection


As explained in the above discussion, battery voltage, battery energy, and the presence of
certain input or output characteristics (e.g., inductive charging) may impact the efficiency of
battery chargers. However, since they also affect the capacity and utility of a charger, DOE must
specify a separate standard level that takes into account any differences in energy consumption
due to differences in these characteristics. Whereas this is straightforward in the case of
inductive chargingi.e., there can be separate product classes with separate standards,
depending on whether a BC uses inductive chargingmatters are more complicated in the case
of battery voltage and battery energy.
Battery voltage depends on the number and chemistry of electrochemical cells in the
battery; while battery energy further depends on the amount of active material in the battery (i.e.,
its capacity). Because the size of the battery is infinitely variable, and the battery can contain
between one and 24 or even more cells, it is possible to establish an arbitrary number of product
classes based on these criteria. While too many product classes would unnecessarily complicate
the analysis and any resultant energy conservation standards, too few product classes may lead to
product classes so large that the BCs that fall inside them have few characteristics in common.
Because DOE conducts its standards analysis by estimating the cost impacts of increasing the
efficiency of a representative unit, it is important that the product classes be delineated such that
the products within the class are similar to the representative unit.
To resolve the question of product class size, DOE examined the topology, or underlying
design, of the power converters that transform input voltage to DC voltage suitable for charging
a battery. The power converter topology determines which technology options can be practically
used to improve the efficiency of a BC. Even though converters of a given topology can vary
depending on capacity and other requirements, many of the technology options will remain
applicable. Basing its product classes on the underlying BC topology therefore allows DOE to
focus its BC standards analysis on a representative unit within each product class and extrapolate
the results for that unit to all products of a similar topology within the class.

3-56

3.3.2.3 Resultant BC Product Classes


DOE first divided BCs into three groups by type of input and output: those with AC input
and DC output, those with DC input and DC output, and those with AC input and AC output.
While many factors influence the choice of topologyincluding experience of the
designer, capabilities of the production facility, time to market, and cost of materials, among
othersoutput power also has a significant effect. Since output power is correlated to battery
energy, DOE researched power converter design guides and manufacturer literature and
evaluated BCs for various applications, in an attempt to generalize the division of topologies by
battery energy. Based on this initial review of topologies, DOE has divided BCs into three
battery energy product classes:
(I) Battery energy less than 100 watt-hours. Most BCs for consumer products charge
batteries smaller than 100 watt-hours and typically rely on line-frequency and flyback
designs. Batteries tend to have lithium-ion or nickel chemistries.
(II) Battery energy greater than or equal to 100 watt-hours and less than 3,000 watthours. BCs that charge batteries in this range tend to use forward and half-bridge
power converter designs. They are used with wheelchair, marine, and lawn mower
applications that rely on sealed lead-acid batteries.
(III) Battery energy greater than or equal to 3,000 watt-hours. BCs that charge
batteries larger than 3,000 watt-hours tend to use ferro-resonant or full-bridge
designs. Their only consumer application is mobilityi.e., golf cars and utility
vehicles, which use flooded lead-acid batteries.
However, battery energy (and therefore topology) is not the only factor that determines
the practicality of technology options that can be used to increase the efficiency of a given BC.
Battery voltage not only constrains which end-use consumer product a given BC can service, as
mentioned above, it also impacts the design of the charger itself. In particular, while certain
technology options may be practical at one voltage, the same may not be true at another voltage,
even within the same topology.
Therefore, for the preliminary analysis, DOE further divided the above battery-energy
based product classes by voltage, dividing the low-energy product class (number I, above) into
low-, medium-, and high-voltage product classes. Similarly, DOE divided the medium-energy
product class (number II, above) into low- and high-voltage product classes. These product
classes along with the others used for the BC preliminary analysis are shown in Table 3.44.

3-57

Table 3.44 BC Product Classes Used in the Preliminary Analysis.


Prod.
Class #
Example Applications
Inductive
1
Toothbrushes
Connection
< 100
0-4 V
2
Telephones
Wh
410 V
3
Cameras and Small Tools
AC In,
1048 V
4
Laptops and Large Tools
DC Out
100
024 V
5
Marine Chargers, Wheelchairs
3,000
2448 V
6
Electric Bikes, Lawnmowers
Wh
> 3000 Wh

3.4
3.4.1

Golf Cars

DC In,
DC Out

5 V Input

USB Chargers

12 V Input

Car Chargers

AC In,
AC Out

AC Output
from Battery

10

Uninterruptible Power Supplies

TECHNOLOGY ASSESSMENT
Introduction

This technology assessment examines BC and EPS technology, with a focus on the
factors affecting their efficiency. It begins by explaining the purpose of EPSs and BCs and their
modes of operation (sections 3.4.2 and 3.4.7). Next, the technology assessment reviews
efficiency metrics and the test procedures established for assessing the performance of EPSs and
BCs in the major energy-consuming modes (sections 3.4.2.1 and 3.4.8). Finally, the assessment
discusses the designs necessary for BCs and EPSs to perform their required function
(sections 3.4.5 and 3.4.9), and the technology options to improve the performance of those
designs against the energy efficiency metrics (sections 3.4.6 and 3.4.10). DOE reviewed these
technology options in the screening analysis and considered them in the engineering analysis to
establish the relationship between cost and efficiency.
3.4.2

EPS Modes of Operation

3.4.2.1 EPS Active Mode


In active mode, the external power supply takes power from mains and converts it to a
form usable by the consumer product or load. For the determination analysis, DOE used the
definition of active mode codified in 10 CFR part 430 subpart B appendix Z: Active mode is the

3-58

mode of operation when the external power supply is connected to the main electricity supply
and the output is connected to a load.
In this mode, EPS efficiency is the conversion efficiency when the load draws some or all
of the maximum rated output power of the EPS. To provide that output power, the EPS also
consumes power due to internal losses as well as overhead circuitry. The amount of power the
EPS consumes varies with the power demands of the load; together, those two parameters define
the EPSs efficiency at a particular loading point:
Pout
Pout
Eq. 3.1
EPS
PIn Pout PEPS _ consumption
EPS efficiency varies with the amount of output power. Typically, EPS efficiency is
lower between 0 and 20 percent of maximum rated output power and higher between 20 and 100
percent of maximum rated output power, where EPSs tend to operate. The lower efficiency at
lower output current is due to the proportionally larger power consumption of internal EPS
components, relative to output power. At higher power, EPS overhead losses increase slightly,
but have less of an effect on EPS efficiency than losses associated with power conversion. The
EPS test procedure evaluates active mode conversion efficiency at four loading points: 25
percent, 50 percent, 75 percent, and 100 percent of maximum rated output power, which captures
a general picture of EPS efficiency. Figure 3.9 shows an example of a typical efficiency curve
for an EPS in active mode.
Figure 3.9 Example of an Efficiency Curve of an EPS in Active Mode
100%

Efficiency [%]

80%

60%

40%

20%

0%
0%

25%

50%

75%

Output current, Iout, as percent of maximum load current, Imax [%]

3-59

100%

3.4.2.2 EPS No-Load Mode


For the determination analysis, DOE used the definition of no-load mode codified in 10
CFR part 430 subpart B appendix Z: No load mode means the mode of operation when the
external power supply is connected to the main electricity supply and the output is not connected
to a load.
EPS efficiency in no-load mode is characterized by EPS power consumption, rather than
conversion efficiency, because the EPS does not deliver power to the load in this mode.
However, the EPS might provide functionality. For example, certain consumer products may
require the EPS to deliver output power within moments of being connected. Thus, the EPS may
consume power to provide the useful function of reduced start-up time. Nonetheless, EPS power
consumption can be low (less than 0.5 watts) in no-load mode.
3.4.3

EPS Efficiency Metrics

An evaluation of the technology options for efficiency improvement and the tradeoffs
between them depends on the metrics used. DOE has previously adopted test procedures for
measuring the energy consumption of both BCs and EPSs.z This section presents a brief
overview of the test procedures for EPSs, and any issues related to the test procedures that may
affect the energy conservation standards rulemaking. See section 3.4.8 for a similar discussion of
BCs.
3.4.4 Energy Efficiency Metrics for External Power Supplies
On December 8, 2006, DOE codified a test procedure final rule for EPSs in Appendix Z
to Subpart B of 10 CFR Part 430 (Uniform Test Method for Measuring the Energy
Consumption of External Power Supplies.) DOEs test procedure, based on the California
Energy Commission (CEC) EPS test procedure, measures active-mode efficiency and no-load
mode (standby-mode) power consumption.
Active mode conversion efficiency is the ratio of output power to input power. DOE
averages the efficiency at four loading conditions25, 50, 75, and 100 percent of maximum
rated output currentto assess the performance of an EPS when powering diverse loads. DOE
also measures the power consumption of the EPS when disconnected from the consumer product,
which is termed no-load power consumption. DOE combines both of the above metrics into
matched pairs that describe the candidate standard levels considered in setting potential energy
conservation standards. This matched pairs combination affected the analysis and is discussed
further in Chapter 5, Engineering Analysis.
3.4.5

EPS Designs

EPSs must meet several specifications in order to power a consumer product; EPSs are
generally designed to provide power at a fixed output voltage with variable current to a
consumer product. The consumer product is what determines the EPS design criteria, including
output power, output voltage and the tolerance of the output voltage. EPSs designed for
consumer products that require precise voltages (e.g., computers) will also incorporate output
z

10 CFR Part 430 Subpart B Appendix Y and Appendix Z

3-60

voltage regulation to minimize voltage fluctuations caused by load or power source variations.
Other applications that can tolerate voltage fluctuation may use simpler EPSs that do not regulate
the output voltage as tightly.
Together, output power and output voltage determine the current, which has the greatest
impact on conduction losses and associated power dissipation in the EPS.
Unregulated and two-stage regulated EPSs are called line-frequency EPSs because the
frequency of the current passing through their transformers is the same as that of the AC mains
current (nominally 60 Hz in the United States). Switched-mode power supplies (SMPS) convert
power differently than line-frequency EPSs. SMPSs first rectify the AC mains voltage to DC,
converting it back to AC by switching the current on and off at high frequency. The highfrequency AC current then passes through the primary winding of a transformer while the output
from the secondary winding of the transformer is rectified, resulting in a low-voltage DC output.
Because of the high frequency of the AC current passing through the transformer, the
transformer can be made smaller, resulting in lower weight, material costs, and losses in the
transformer.
3.4.5.1 AC/AC External Power Supplies
An AC/AC external power supply is the simplest type of EPS, typically consisting only
of a transformer. A transformer contains two wires wrapped around a metal core; as current
passes through the primary wire, power is transferred to the secondary wire (usually at a lower
voltage) through magnetic induction in the core. The induced voltage depends on the relative
number of turns between the primary and the secondary wires. The windings of the transformer
are wound so that the voltage generated in the secondary wire is at the design voltage for the
consumer product when mains voltage is applied to the primary wire. Because the primary and
secondary windings are two separate wires, the transformer also provides a safety function,
electrically isolating the consumer product from mains. The key factors that determine
transformer losses are core size, core material, number of windings, and wire gauge.
Transformer
Primary

Secondary

VMains

VOut

Figure 3.10 Circuit Diagram for an AC/AC External Power Supply

3-61

3.4.5.2 Unregulated Line-Frequency AC/DC External Power Supplies


In unregulated line-frequency EPSs, the two main sources of loss are the transformer and
the rectifying diodes. After passing through the transformer, current passes through rectifier
diodes, which have voltage drops that also dissipate power. Typically, diodes have a drop of 0.6
volts, which constitutes a proportionally larger share of the losses at lower output voltages. For
AC-DC EPS that have a low output voltage, below approximately 12 V the power consumed by
the diodes also becomes significant. A line-frequency raw supply has three distinct stages
(Figure 3.11): a transformer to isolate and step down mains voltage, a rectifier to convert AC
voltage to DC voltage, and a filter capacitor to smooth the output voltage.

Transformer
Primary

Rectifier

Filter

Secondary

VMains

VOut

Figure 3.11 Circuit Diagram of a Line-Frequency Raw Supply


The raw supply, consisting of a transformer, rectifier, and filter capacitor, is directly
responsive to the load. A change in mains power or the resistance of the load directly affects the
output voltage of the raw supply. If required, a regulator circuit follows the raw supply circuit,
housed either in the EPS or in the end-use product before the load.
3.4.5.3 Linear-Regulated Line-Frequency AC/DC External Power Supplies
To achieve voltage regulation, manufacturers can add a second stage, such as a linear
regulator, to the line-frequency power conversion stage described above, or redesign the power
conversion stage entirely using a switched-mode topology. Of the two regulator technologies,
linear regulators are simpler, bulkier, cheaper, and generally less efficient at higher power levels
than switching regulators. Switching regulators, although more complicated and costly, provide a
good alternative when portability or over-heating is a concern, such as when an EPS is used with
a mobile phone charger or a high-power flat-panel television.
The AC-DC conversion stage of a regulated line-frequency EPS is essentially the same as
that of an unregulated EPS, with the same sources of power consumption. The linear voltage
regulation stage adds to these losses by passing power from the AC-DC converter to the
consumer product through a power-dissipating element. This regulation stage senses the output
voltage and adjusts the voltage across it to keep the output voltage proportional to a fixed
reference voltage. Loss in a regulated line-frequency EPS is caused by the conversion stage
delivering current at a higher voltage than needed by the consumer product, and dropping the

3-62

excess voltage across the regulator to achieve the lower regulated output voltage. Dissipated as
heat, the power lost in the regulator is the product of the voltage drop and the load current.
Linear regulators have two key elements: a sensor and a pass device, which work
together to produce a fixed output voltage (Figure 3.12). To determine those adjustments, the
sensor element continuously compares the output voltage to a reference voltage. Whenever there
is a difference between the two voltages, the sensor directs the pass device to adjust the output in
order to reduce that difference. This continuous adjustment allows the regulator to yield a
constant output voltage as the load resistance or mains voltage varies. The output voltage of the
linear regulator circuit is what the user sees as the output voltage of the EPS.

3-63

Linear Regulator
Stage
Mains
Power

Raw Supply
Stage

Pass Device

VOut

Sensor

Figure 3.12 Block Diagram of a Linear Regulator


Figure 3.13 shows a circuit diagram of a low-dropout linear regulator, one of the more
common types of linear regulators. To determine that voltage drop, an operational amplifier
(commonly referred to as an op-amp) acts as a sensor that compares the output voltage against
a reference voltage. Based on those two signals, the op-amp controls a transistor, which is the
pass device. The voltage drop across the transistor determines the output voltage but also
dissipates energy. The energy dissipated by the pass device is the main source of energy
consumption in the linear regulator, and hence the main source of inefficiency and heat
generation. Together, the sensor and the pass device adjust the output of the regulator to produce
a relatively stable output voltage, which is what the load receives as the output voltage of the
EPS.
Heat Dissipated

Pass
Device

+
+

VRawSupply_out

VReference

Op
Amp
-

VOut

Sensor

Figure 3.13 Simplified Circuit Diagram of a Linear Regulator

The efficiency of the linear regulator, LinReg, is:

3-64

LinReg

PLinReg _ out

VLinReg _ out I LinReg _ out

PLinReg _ in

VLinReg _ in I LinReg _ in

Eq. 3.2

where PLinReg_out is the power out of the linear regulator,


PLinReg_in is the power into the linear regulator,
VLinReg_out is the voltage out of the linear regulator,
VLinReg_in is the voltage into the linear regulator,
ILinReg_out is the current out of the linear regulator, and
ILinReg_in is the current into the linear regulator.

Because the linear regulator connects to the raw supply, VLinReg_in is equal to VRawSupp_out,
the output voltage of the raw supply. Furthermore, because the input current flows directly to the
output through the pass device, with other currents being negligible, I LinReg _ out I LinReg _ in .
Therefore, the efficiency of the linear regulator alone is approximately:
VLinReg _ out
LinReg

Eq. 3.3

VLinReg _ in

The total efficiency of an EPS with a linear regulator depends on the efficiency of both
the linear regulator stage and the raw supply stage. Depending on the load conditions, LinReg
generally ranges from 0.6 to 0.8, meaning the linear regulator is about 60 to 80 percent efficient.
The efficiency of the raw supply, RawSup, also varies with the load, but is generally from 0.7 to
0.9. The raw supply and linear regulator each are most efficient at different load conditions.
Multiplied, LinReg and RawSupp yield the total efficiency of an EPS with a linear regulator,
Lin_EPS, which is generally about 50 percent, but is lower for EPSs with output power below
10 W:
Lin _ EPS

RawSupp

LinReg

Eq. 3.4

For an EPS consisting of a raw supply and a linear regulator, mains voltage at line
frequency (60 Hz) is directly applied to the transformer. If the power applied to the transformer
had similar voltage and current characteristics but a higher frequency, the transformer could be
smaller and lighter. Those benefits are part of the motivation for choosing switching regulators,
which, unlike their linear counterparts, have transformers that operate at high frequency (greater
than 20 kHz).
3.4.5.4 Switching-Regulated Line-Frequency AC/DC External Power Supplies
A switching regulator can also follow the line-frequency AC-DC power-conversion stage
in place of the linear regulator described above, which is different from the switched-mode EPS
discussed below. These tend to be much more efficient than linear regulators because they do not
dissipate excess power through a linear control element. Rather, they switch the current at high
frequency, adjusting the proportion of on time during each switching cycle (i.e., the duty ratio) to

3-65

maintain the regulated output voltage proportional to a fixed reference. Due to their higher costs,
these switching regulators tend not to be as common as linear regulators.
Switching Regulator Stage

Mains
Power

Raw Supply
Stage

Energy
Transfer

Chopper

Control

VOut

Feedback
Isolation

Figure 3.14 Block Diagram of a Switching-Regulated Line-Frequency AC/DC EPS


3.4.5.5 Switched-Mode AC/DC External Power Supplies
The most common method for regulating power to a consumer product is a switchedmode EPS. The critical elements in a switched-mode EPS are the transistor, output rectifier, bulk
capacitor, transformer, and controller. A transistor acts as a switch that constrains the flow of
power rectified from mains into the transformer (or choke), through the output rectifier, and,
ultimately, to the consumer product. A controller, typically an integrated circuit, switches the
transistor on and off based on the output voltage. By adjusting the duty ratio, the IC controls the
rectified mains current into the primary winding of the transformer and thereby the output
voltage of the EPS. The IC can also limit power dissipation in active mode by switching at low
current or low voltage. Further, the IC can greatly increase efficiency by reducing power
consumption in no-load mode, the condition when the EPS has been disconnected from the load,
resulting in zero output current. After passing through the transformer, the current is rectified and
filtered before reaching the consumer product. Principal sources of loss in a switched-mode EPS
are the transistor switching transients, magnetization and resistive losses as a result of
transformer current, controller IC power consumption, and rectifier losses. Although there are
more sources of loss for switched-mode EPSs than line-frequency EPSs, in total, losses in
switched-mode EPSs tend to be lower.
The switching regulator consists of five stages: an AC-DC conversion stage, a chopper
stage, an energy transfer stage, a control stage, and a feedback isolation stage (Figure 3.15).
First, the current is rectified and passed to the chopper, which converts the DC voltage back to
AC, but at high frequency. The energy transfer stage then takes energy from the chopper, briefly
stores it, and then passes it to the rectifier to be output to the consumer product. The energy
transfer stage also serves to isolate the user from the mains. The level of the output voltage is fed
back through an isolation stage to the controller, which tracks the output voltage and adjusts the
chopper to make the desired voltage.

3-66

Switched-mode Power Supply


Mains

AC/DC
Conversion

Chopper

Control

Energy
Transfer

VOut

Feedback
Isolation

Figure 3.15 Block Diagram of a Switched-Mode Power Supply


The switching regulator usually consists of an integrated circuit controller and discrete
components. The circuit diagram in Figure 3.16 depicts a flyback switching regulator, one of
the more common types; however, many other switching regulator designs also exist. The
AC/DC conversion stage consists of a diode bridge and filter capacitor, similar to a raw supply.
In this case, current flows directly from mains to the diode bridge, rather than through a
transformer.
The chopper stage uses a transistor, which switches on and off at high frequency to
convert the DC current from the AC/DC converter back to an AC current for the energy transfer
stage. A control stage drives the transistor, where the longer its on time in the duty cycle, the
more energy is transferred. The switching frequency is in the kilohertz range, with lower
frequencies having lower switching losses. Typically, the minimum frequency is 20 kHz, above
the audible range of human hearing.

3-67

Energy Transfer

AC/DC Conversion

VOut

VMains
-

Chopper

PWM

Optocoupler

Control

Isolation

Figure 3.16 Simplified Circuit Diagram of a Flyback Switching Regulator


The energy transfer stage consists of a choke, a capacitor, and a diode. The choke is
similar to a transformer and has the same symbol in the circuit diagram. One difference is that a
transformer in a line-frequency EPS is designed to pass energy from one winding to another with
minimal energy storage, while a choke in a switched-mode EPS is designed to store and release
energy. Also, the phasing of the windings is not important in the line-frequency transformer, but
it is critically important in the flyback switching regulator. This is represented in Figure 3.16 by
dots on the choke.
When the chopper switch is closed, the primary winding of the choke takes energy from
the chopper and stores the energy in the choke. When the chopper switch opens, the secondary
winding transfers that energy through the diode to the capacitor and provides the output for the
switching regulator, electrically isolating the load from the mains. Because the choke operates at
a high frequency, it benefits from the associated decreases in size and weight. The energy
transfer scheme of the switching regulator is more efficient than a linear regulator, in part
because the choke stores and returns energy with relatively low losses.
The isolation stage typically uses an optocoupler that consists of a light source and a
photosensitive detector. By converting the electrical feedback signal to an optical one, the
optocoupler maintains the load electrically isolated from the mains. The detector converts the
optical signal back to an electrical signal that it provides to the controller.
Generally, the controller is an integrated circuit that drives the chopper with a highfrequency pulse-width-modulated (PWM) waveform. The controller monitors the EPS output
voltage and adjusts the pulse width to increase or decrease the amount of energy transferred by
the chopper. If the output voltage dips, the controller will increase the duty cycle, thus increasing
the energy passed by the energy transfer stage and increasing the output voltage. Conversely, if
the output voltage rises, the controller will decrease the duty cycle or possibly skip cycles. This
cycle-skipping feature is especially useful when there is no load attached, because the EPS will
only take from the mains the small amount of power it needs to power itself.

3-68

Instead of an IC controller, a switched-mode EPS can also use discrete components, as in


the case of a ringing choke converter. In that topology, discrete transistors control the chopper
and the resulting energy transfer. EPSs without ICs tend to be more common at lower output
powers. However, without an IC, an EPS cannot offer functions such as cycle-skipping at low
load, which can be used to decrease losses.
Both linear-regulated and switching-regulated EPS use regulating circuits to achieve a
stable output voltage. However, voltage is not the only output variable that can be regulated.
Current regulation, as discussed in the following section, is a fundamental consideration in BC
design.
3.4.6

EPS Technology Options


DOE considered seven technology options that may improve the efficiency of EPSs:
Improved Transformers. In line-frequency EPSs, the transformer has the largest effect
on efficiency. Transformer efficiency can be improved by replacing their cores and
windings with ones made of lower-loss material or adding extra material.
Switched-Mode Power Supply. Line-frequency EPSs may use linear regulators to
maintain a constant output voltage. By using a switched-mode circuit architecture, a
designer can limit both losses associated with the transformer and the regulator. The
differences between the two EPS types are discussed in section 3.4.5.3 and section
3.4.5.4.
Low-Power Integrated Circuits. The efficiency of the EPS can be further improved
by substituting low-power IC controllers, which can switch more efficiently in active
mode and reduce power consumption in no-load mode. For instance, the IC can turn
off its start-up current (sourced from the primary side of the power supply) once the
output voltage is stable. In addition, when in no-load mode, the IC can turn off the
switching transistor for extended periods of time (termed "cycle-skipping").
Schottky Diodes and Synchronous Rectification. Both line-frequency and switchedmode EPSs use diodes to rectify output voltage. Schottky diodes and synchronous
rectification can replace standard diodes to reduce rectification losses, which are
increasingly significant at low voltage. Schottky diodes have a voltage drop of 0.3
0.4 volts, compared to approximately 0.6 volts for standard diodes. Synchronous
rectification (typically only used in switched-mode EPSs) further reduces losses by
substituting transistors for the diodes. The voltage drop across the drain-to-source
resistance of transistor is much lower than that across even a Schottky diode, leading
to lower losses in the output rectifier.
Low-Loss Transistors. The switching transistor dissipates energy due to its drain-to
source resistance (RDS_ON) when the current flows through the transistor to the
transformer. Using transistors with low RDS_ON can reduce this loss.
Resonant Switching. In addition to reducing the RDS_ON of the transistor, power
consumption can be lowered further by the IC controller decreasing switching
3-69

transients through zero-voltage or zero-current switching. The power consumption of


the transistor is influenced by the voltage across the RDS_ON and the current flowing
through it. An IC can control the switching to minimize that voltage or current,
although some components in addition to the IC may also be needed.
Resonant ("Lossless") Snubbers. In switched-mode EPSs, a common snubber protects
the switching transistor from the high voltage spike that occurs after the transistor
turns off by dissipating that power as heat. A resonant or lossless snubber recycles
that energy rather than dissipating it.

3.4.7

BC Modes of Operation

Like the design of EPSs, the design of BCs is driven by the anticipated power
requirements and time spent in their various modes of operation. Section 325(gg)(1)(A) of
EPCA, as modified by EISA, defines active, standby, and off modes for consumer products in
general. (42 U.S.C. 6295(gg)(1)(A)) However, section 2 of appendix Y to 10 CFR part 43046
(hereafter referred to as appendix Y) defines additional modes as well as redefines some of the
EISA modes to be more applicable to BCs (as allowed under 42 U.S.C. 6295(gg)(1)(B))
3.4.7.1 Active or Charge Mode
Active mode is defined as the condition in which an energy-using product(I) is
connected to a main power source; (II) has been activated; and (III) provides 1 or more main
functions. (42 U.S.C. 6295(gg)(1)(A)(i)) However, paragraph 2.i of appendix Y further
specifies that the charger is in active mode specifically when charging a depleted battery,
equalizing its cells,aa or performing other one-time or limited-time functions necessary for
bringing the battery to the fully charged state.
3.4.7.2 Maintenance Mode
Once the batteries have reached full charge, the BC typically enters a maintenance mode,
intended to maintain the fully charged state of the battery, while protecting it from overcharge.
BCs without a maintenance mode (some high-power BCs for consumer motive equipment, for
example) either use a timer to disconnect the BC from the batteries after charging or rely on the
user to manually disconnect.
3.4.7.3 Standby or No-Battery Mode
Alternatively, following a full charge, the user can remove the battery (or in the case of
integral-battery products, the end-use product and the battery), placing the battery charger in
standby or no-battery mode. Typically, the BC is in the mode when the application it serves is in
use; however, the user may also place the BC in off mode, or disconnect it from mains entirely.

aa

Equalization serves to balance the voltage across each of the cells in a multi-cell battery, a process that is most
commonly performed with large lead-acid batteries. Unbalanced cells limit charge and discharge, reducing the
usable capacity; they can also suffer more overcharge than the other cells.

3-70

3.4.7.4 Other Modes and Applicability


Appendix Y defines standby or no-battery mode as the condition in which (1) the battery
charger is connected to the main electricity supply; (2) the battery is not connected to the
charger; and (3) for battery chargers with manual on-off switches, all such switches are turned
on. However, if (1) the charger remains connected to mains, (2) the battery is removed, and (3)
all manual on-off switches are turned off, the charger is then placed in off mode.
Because it has purposely been disabled by the user via a switch, the BC must no longer
perform standby-mode functions such as powering circuitry that detects the presence of a battery
or indicates its status. It therefore has the potential to consume less energy than in standby mode.
Finally, the user can also disconnect the charger from mains, in which case it does not consume
any energy.
Whether each of the modes described above apply to a particular BC depends on whether
the battery is integral or detachable, the presence of manual on-off switches, etc. For example,
BCs without a manual on-off switch cannot be placed in off mode, while a BC with a nonremovable AC cord and integral batteries that are not removed from the application for charging
cannot be placed in no-battery mode.
Nonetheless, all BCs operate in the active or charge mode by definition. This mode has
the largest effect on the BCs size and efficiency because the charger must be designed to
accommodate the maximum amount of power output, which happens during active mode. While
the requirements of the other modes factor into the design as wellas does the chemistry of the
batterytheir effects on efficiency are not as great, since they dont affect the power handling
components, but rather sub-circuits tasked with assessing the state of charge and ensuring safety.
3.4.8

BC Efficiency Metrics

On December 8, 2006, DOE adopted a test method to measure the efficiency of battery
chargers. 71 FR 71340. This test method, based on the U.S. Environmental Protection Agencys
(EPA) ENERGY STAR Test Methodology for Determining the Energy Performance of Battery
Charging Systems, integrates the power consumed by BCs in maintenance and no-battery
modes over fixed periods of time. This non-active energy is divided by the battery energy,
measured at a discharge rate of 0.2 C, resulting in an energy ratio. Normalizing by battery energy
is meant to account for proportionally higher losses in chargers intended for higher-energy
batteries. A higher energy ratio represents higher BC non-active energy consumption.
However, in the December 8, 2006, Test Procedure Final Rule, DOE stated that it
intended to study further BC active mode energy consumption and reserved a section in the test
procedure (section 4(b) of appendix Y to subpart B of 10 CFR Part 430) to cover measurement
of active mode energy consumption. 71 FR 71340, 71360. DOE has continued developing its
approach for measuring BC activei.e., chargingmode energy consumption and on April 2,
2010published a proposal for adopting an active mode test procedure based on a test procedure
previously adopted by the California Energy Commission. 75 FR 16958.
If adopted, this proposed amendment would stop use of the non-active energy ratio in
favor of metrics corresponding to energy consumption in each of the energy-consuming modes

3-71

of operation of a BC. As described above, these include active, maintenance, standby, and off
modes; thus, the test procedure would return four separate metrics. These separate metrics would
then be weighted by an average usage profilebb that reflects the typical usage of BCs in each
product class. For the analysis in this preliminary TSD, DOE used these metrics to evaluate BC
efficiency.
The potential energy conservation standards for each class would likewise be written in
terms of a single metric, even though the test procedure would measure consumption in each of
the modes separately. Manufacturers would then be free to trade off power consumption in one
mode for that in another, as long as they meet the usage-weighted energy consumption required
by the standard.
3.4.9

Battery Charger Design

The design of a battery charger depends on the application it serves, and as mentioned in
the discussion of product classes in section 3.3, specifically its voltage and energy requirements.
As a result, the design of battery chargers varies with product class, which is defined by battery
voltage and energy. Therefore, following a brief introduction, this section will be divided by
product class.
A general schematic of a battery charging system (BC and battery) can be seen in Figure
3.17. As indicated in the figure, the primary function of a BC is regulating the flow of current
from a power supply to a battery to safely charge the battery and maintain its charge.
Rechargeable
Battery

Battery Charger
Power Supply
IOut

RSupply

+
+
+
-

VSupply

Charge
Regulator

VBatt
-

Electrochemical
Process
-

Figure 3.17 General schematic of a BC and battery.


As depicted in Figure 3.17, the first stage of most BCs is a power supply that converts
line-voltage AC power to DC power at a voltage low enough to charge the battery (through a
charge regulator) as well as power any overhead circuitry in the charger responsible for safety
and other user function (e.g., status indicators, etc.; not pictured in the figure). Because of this
general AC/DC conversion requirement, the design of the input power supply stage often mimics
bb

The calculation of product-class-average usage profiles and the calculation of typical energy consumption is
described in detail in the energy use analysis (chapter 7).

3-72

that of the EPSs described in section 3.4.5, in particular unregulated line-frequency AC/DC and
various switched-mode AC/DC designs, including flyback. This power supply stage can be either
internal (i.e., in the same enclosure as the rest of the BC circuitry) or in an external wall adapter.
Although the circuits are similar, the key difference between an EPS and a BC is charge control.
An EPS is designed to provide output current so as to maintain a constant output voltage to the
load. In contrast, a BC provides power to a battery that may either be constant current or constant
voltage, depending on the state of the battery. Further, a battery sets a voltage in the BC circuit,
whereas a load does not set a voltage for an EPS.
Methods of improving BC efficiency depend on whether the BC is a slow charger or a
fast charger. The distinction between the two types of BCs is based on the charge rate (also
referred to as C-rate), often defined as the average charging current flowing into the battery,
divided by the nominal battery charge capacity. For current expressed in amperes and battery
capacity expressed in ampere-hours, the resulting quantity is expressed in units of 1/hours or C.
For example, a BC with a 0.1 ampere (A) output current charging a 1 ampere-hour (Ah) battery
would result in a charge rate of 0.1 C. Charging time is approximately the inverse of the charge
rate, adjusted for the efficiency of the battery itself, which varies with chemistry. In the previous
example, the battery would take slightly longer than 10 hours to charge.
DOE considers BCs with charge rates less than 0.2 C (typically around 0.1 C) to be slow
chargers. At this low charge rate, nickel-based batteries can be charged continuously without
concern for excessive battery overheating or safety. Slow chargers do not typically include cutoff
or monitoring circuitry. However, as the battery nears full charge and its voltage increases, the
difference between the BC output and battery voltages decreases and the charge-control
resistance used in a slow charger will cause the charging current to decrease. This reduces power
consumption and lessens battery heating due to overcharge (thereby extending battery life). Slow
chargers are not typically used in combination with lithium-based batteries, because of the safety
concerns associated with overcharging lithium-based batteries.
Slow chargers are typically composed of a line-frequency transformer followed by a
rectifier and charge-control element. The function of the charge-control element is to limit
charging current into the battery, which can be accomplished by either a discrete resistor or the
parasitic internal resistance of the transformer windings. The power conversion losses in a slow
charger are mostly due to magnetization losses in the transformer core steel, resistive losses in
the charge-control element, and voltage drops across the rectifier diodes.
In addition, slow chargers typically continue to deliver current to the battery even after it
is fully charged, usually at a rate much higher than that necessary to maintain the charge lost due
to battery self-discharge. The excess power is dissipated as heat in the battery. The power
conversion losses in the BC identified earlier continue to have an impact in this maintenance
mode, further increasing power consumption. Even in no-battery mode, when the battery is
disconnected from the charger, the slow charger continues to consume significant power due to
the transformer magnetization losses. For a detailed discussion of slow-charger power
consumption in all modes, please see sections 3.3 and 3.5 of the draft technical report that
accompanied the Framework Document published on June 4, 2009. 74 FR 26816.

3-73

A battery charger that contains monitoring, cutoff, or limiting circuitry can safely charge
lithium-based batteries and fast-charge nickel-based batteries. DOE considers BCs with charge
rates greater than 0.2 C (typically between 0.6 C and 1 C) to be fast chargers. Because the charge
rate of fast chargers is much greater than that of slow chargers, the maximum rated output power
of a fast charger can be 5 to 20 times greater than that of slow chargers, even when charging a
battery of the same voltage and capacity. For this reason, fast chargers typically use switchedmode power supplies, which are smaller and lighter than line-frequency power supplies. Fast
chargers also employ monitoring and cutoff circuitry, as the high currents used during charging
may overheat the battery and lead to a safety hazard if not reduced at the proper time. Because of
these design differences, fast chargers are composed of more complex circuits and are
susceptible to different loss mechanisms than slow chargers.
The high-frequency switched-mode power supply (whether internal or external) that
typically performs the energy conversion in a fast charger is usually more efficient than the linefrequency transformer and rectifier discussed previously. High-frequency power supplies can use
transformer cores made of ferrite that are smaller and more efficient than the steel cores typically
found in line-frequency designs. However, there are still conversion losses associated with
switching and rectification, as well as fixed overhead losses associated with powering the
integrated circuit (IC) switching controller and any safety circuitry. Also, although fast chargers
terminate (i.e., limit charging current once the battery has reached full charge), most chargers
continue to supply a small amount of maintenance current. As with slow chargers, this
maintenance current and the associated conversion losses contribute heavily to maintenancemode power consumption. Finally, even with the battery removed, the charger can continue
consuming significant power due to the overhead of powering the control and safety circuitry
mentioned above. For a more detailed discussion of fast-charger power consumption, please see
sections 3.3 and 3.5 of the draft technical report.
Further, manufacturers may, and often do, choose to substitute a fast charger for a slow
one as a means of improving portability and energy efficiency. Because both types of chargers
can often be used with the same battery powering the same consumer product, they provide the
same utility to the consumer, which sometimes means the fast charger can be considered as a
replacement for a slow charger.
Finally, because changes in battery temperature and voltage happen more slowly at lower
charge rates, monitoring circuitry that depends on these changes to stop the charging process is
typically not sensitive enough to be used at rates below 0.3 C. Therefore, although DOE
differentiates between BCs with charge rates greater than or less than 0.2 C, DOE does not
expect to find many BCs with charge rates between 0.15 C and 0.3 C.
3.4.10 Battery Charger Technology Options
Battery charger efficiency in active mode is governed by BC component losses and
overhead circuitry. BCs share with EPSs similar options for reducing component losses in active
mode. However, some BCs have safety circuitry to monitor the battery during charging, which
EPSs typically do not include. Safety circuits are often present in BCs that are fast chargers;
safety concerns also affect design of slow charging BCs. Thus, if a BC were compared to an EPS
with similar power ratings, it might appear to have lower conversion efficiency due to the
additional power consumption of its safety circuitry.
3-74

The following list, organized by charger type, provides preliminary technology options
that DOE intends to evaluate during the preliminary analyses. A detailed discussion of the
specific technology options can be found in section 3.8.2 of the draft technical report. Although
many of these technology options could be used in both fast and slow chargers, doing so may be
impractical due to the cost and benefits of each option for the two types of chargers. Therefore,
in the list below, the options are grouped with the charger type where they would be most
practical.
Slow charger technology options include:
Improved Cores: The efficiency of line-frequency transformers, which are a
component of the power conversion circuitry of many slow chargers, can be
improved by replacing their cores with ones made of lower-loss steel.
Termination: Substantially decreasing the charge current to the battery after it has
reached full charge, either by using a timer or sensor, can significantly decrease
maintenance-mode power consumption. Because most slow chargers have a charge
rate of approximately 0.1 C, and maintenance-mode current below 0.05 C is typically
sufficient to keep a battery fully charged, a slow charger that employs termination can
roughly halve its maintenance-mode power consumption.
Elimination/Limitation of Maintenance Current: Constant maintenance current is not
required to keep a battery fully charged. Instead, the BC can provide current pulses to
"top off" the battery as needed. Elimination or limitation of maintenance can decrease
maintenance-mode power consumption even further and has the added benefit of
extending the battery lifetime by reducing heating due to overcharge.
Elimination of No-Battery Current: A mechanical AC line switch inside the battery
charger "cup" automatically disconnects the BC from the mains supply when the
battery is removed from the charger. Although manual (i.e., user-controlled) switches
are also possible, this method guarantees that the BC ceases to consume power once
the battery is removed from the battery charger.
Switched-Mode Power Supply: To increase efficiency, line-frequency power supplies
can be replaced with switched-mode EPSs, which greatly reduce the biggest sources
of loss in a line-frequency EPS: the transformer. Because a switched-mode EPS
operates at high frequency (greater than 20 kHz), its transformer can be smaller, and
because transformer losses are a function of volume, a smaller transformer is usually
more efficient. It is worth noting that this technology option is not often found in
practice, because the inclusion of a switched-mode power supply within the BC
design allows the higher power levels necessary for fast charging. The universal
consumer preference for shorter charging times limits the occurrence of slow chargers
with high-frequency switched-mode power supplies.
Fast charger technology options include:

3-75

Low-Power Integrated Circuits: The efficiency of the BC's switched-mode power


supply can be further improved by substituting low-power IC controllers, which can
switch more efficiently in active mode and reduce power consumption in no-load
mode. To increase efficiency in active mode, the IC controller can decrease switching
transients through zero-voltage or zero-current switching. Furthermore, the IC can
turn off its start-up current (sourced from the primary side of the power supply) once
the output voltage is stable. In addition, when in no-load mode, the IC can turn off the
switching transistor for extended periods of time (termed "cycle-skipping").
Elimination/Limitation of Maintenance Current: See above.
Schottky Diodes and Synchronous Rectification: Both line-frequency and switchedmode EPSs use diodes to rectify output voltage. Schottky diodes and synchronous
rectification can replace standard diodes to reduce rectification losses, which are
increasingly significant at low voltage. Schottky diodes are rectifiers constructed
from a metal-silicon junction rather than a p-n silicon junction and have a voltage
drop of 0.30.4 volts, compared to approximately 0.6 volts for standard p-n junction
diodes. Synchronous rectification (which is typically used only in switched-mode
EPSs) further reduces losses by substituting field-effect transistors (FETs) for the
diodes. The voltage drop across the drain-to-source resistance of the FET is much
lower than that of a Schottky diode, leading to lower losses in the output rectifier.
Elimination of No-Battery Current: See above.
Phase Control to Limit Input Power: Even when a typical BC is not delivering its
maximum output current to the battery, its power conversion circuitry continues to
draw significant power. A phase control circuit, like the one present in most common
light dimmers, can be added to the primary side of the BC power supply circuitry to
limit input current in lower-power modes.

3-76

REFERENCES

Consumer Electronics Association Market Research Department. U.S. Consumer Electronics Sales and Forecasts

2005-2010. July 2009. Consumer Electronics Association: Arlington, VA.

2
AppleInsider.com. Apple snags 10 percent of U.S. retail notebook sales in March. May 11, 2007. (Last accessed

September 13, 2010.)

<http://www.appleinsider.com/articles/07/05/11/apple_snags_10_percent_of_u_s_retail_notebook_sales_in_march.

html.> The September 13, 2010 material from this website is available in Docket # EERE-2008-BT-STD-0005. For

more information, contact Ms. Brenda Edwards at (202) 586-2945.

3
RocketFish. Digital Wireless Speakers. 2008. Available at:

<http://www.rocketfishproducts.com/cms/documents/RF-WS01_WS02_WS02-W_08-1669_ENGLISH_WEB.pdf>

4
U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy. FY2005 Preliminary Priority-

Setting Summary Report and Actions Proposed. Appendix A: FY2005 Technical Support Document. Table A9-1:

Background Data on Battery Chargers and Power Supplies.

5
Whitney, Lance. Netbook sales to rise as notebooks fall flat. CNET News. July 13, 2009. (Last accessed July 15,

2010.) <http://news.cnet.com/8301-1001_3-10285189-92.html?tag=mncol> The July 15, 2010 material from this

website is available in Docket # EERE-2008-BT-STD-0005. For more information, contact Ms. Brenda Edwards at

(202) 586-2945.

6
EE Times Asia. HDD shipments hit 15.5% growth amid challenges. 2007. (Last accessed September 13, 2010.)

<http://www.eetasia.com/ART_8800456083_499486_NT_67de3825.HTM> The September 13, 2010 version of this

website is available in Docket # EERE-2008-BT-STD-0005. For more information, contact Ms. Brenda Edwards at

(202) 586-2945.

7
World Bank. Gross domestic product 2008. 2009. Available at:

<http://siteresources.worldbank.org/DATASTATISTICS/Resources/GDP.pdf>

8
Global Industry Analysts. Uninterruptible Power Supply Systems. May 2008. Report MCP-1120.

9
Appliance Magazine. May 2009.

10
Masterson, Michele. Printer Market Q4 2008: News That Isn't Fit To Print. ChannelWeb. March 16, 2009. (Last

accessed September 13, 2010.)

<http://www.crn.com/hardware/215900485;jsessionid=M1XEH1PUFGCELQE1GHPCKH4ATMY32JVN> The

September 13, 2010 version of this website is available in Docket # EERE-2008-BT-STD-0005. For more

information, contact Ms. Brenda Edwards at (202) 586-2945.

11
Cox, Mark. Printer market shows strong sequential growth: IDC. eChannelLine. December 6, 2009. (Last

accessed September 13, 2010.) <http://www.echannelline.com/usa/story.cfm?item=25239> The September 13, 2010

version of this website is available in Docket # EERE-2008-BT-STD-0005. For more information, contact Ms.

Brenda Edwards at (202) 586-2945.

12
Darnell Group, Inc. External AC-DC Power Supplies Worldwide Forecasts Third Edition. May 2008. North

America supplement prepared for DOE.

13
RCR Wireless News. Bluetooth adoption set for continued expansion. 2009. (Last accessed September 13,

2010.) <http://www.rcrwireless.com/article/20090119/WIRELESS/901199989/1080/bluetooth-adoption-set-for
continued-expansion> The September 13, 2010 version of this website is available in Docket # EERE-2008-BT
STD-0005. For more information, contact Ms. Brenda Edwards at (202) 586-2945.

14
IT Facts. 240 mln GPS-enabled phones to sell in 2009. 2009. (Last accessed September 13, 2010.)

<http://www.itfacts.biz/240-mln-gps-enabled-phones-to-sell-in-2009/12494> The September 13, 2010 version of

this website is available in Docket # EERE-2008-BT-STD-0005. For more information, contact Ms. Brenda

Edwards at (202) 586-2945.

15
IT Facts. 500 mln music phones shipped in 2007. 2008. (Last accessed September 13, 2010.)

<http://www.itfacts.biz/500-mln-music-phones-shipped-in-2007/10294> The September 13, 2010 version of this

website is available in Docket # EERE-2008-BT-STD-0005. For more information, contact Ms. Brenda Edwards at

(202) 586-2945.

16
GSM World. Universal Charging Solutions. 2009. (Last accessed September 13, 2010.)

<http://www.gsmworld.com/our-work/mobile_planet/universal_charging_solution.htm> The September 13, 2010

version of this website is available in Docket # EERE-2008-BT-STD-0005. For more information, contact Ms.

Brenda Edwards at (202) 586-2945.

3-77

17

IT Facts. Americans discard 125 mln phones annually. 2008. (Last accessed September 13, 2010.)
<http://www.itfacts.biz/americans-discard-125-mln-phones-annually/10093> The September 13, 2010 version of
this website is available in Docket # EERE-2008-BT-STD-0005. For more information, contact Ms. Brenda
Edwards at (202) 586-2945.
18
GSM World. Mobile Industry Unites to Drive Universal Charging Solution for Mobile Phones. 2009. (Last
accessed September 13, 2010.) <http://www.gsmworld.com/newsroom/press-releases/2009/2548.htm> The
September 13, 2010 version of this website is available in Docket # EERE-2008-BT-STD-0005. For more
information, contact Ms. Brenda Edwards at (202) 586-2945.
19
Shambora, Jessica. Consumers Flip for Mini Camcorders. CNN Money. January 27, 2009. (Last accessed
September 13, 2010.) <http://money.cnn.com/2009/01/27/technology/flip_cams.fortune/> The September 13, 2010
version of this website is available in Docket # EERE-2008-BT-STD-0005. For more information, contact Ms.
Brenda Edwards at (202) 586-2945.
20
Clairmont, Kristy. PMA Data Watch: Purchasing of cameras and accessories. April 2009. (Last accessed
September 13, 2010.) <http://pmaforesight.com/?p=155> The September 13, 2010 version of this website is
available in Docket # EERE-2008-BT-STD-0005. For more information, contact Ms. Brenda Edwards at (202) 586
2945.
21
56th Annual Report: Statistical Review. Appliance Magazine. 2009. 66 (4): pp 34-37.
22
iRobot Corporation. 2008 iRobot Annual Report. 2009. Bedford, MA. Report No. 000-51598
23
Kaiser, Sheryl. Child Safety Gate. (Last accessed July 15, 2010.)
<http://www.firstbabymall.com/product/gate.htm> The July 15, 2010 version of this website is available in Docket #
EERE-2008-BT-STD-0005. For more information, contact Ms. Brenda Edwards at (202) 586-2945.
24
Orland, Kyle. February NPD data shows DS, Wii on top. March 16, 2007. (Last accessed July 14, 2010.)
<http://www.joystiq.com/2007/03/16/february-npd-data-show-ds-wii-on-top/> The July 14, 2010 version of this
website is available in Docket # EERE-2008-BT-STD-0005. For more information, contact Ms. Brenda Edwards at
(202) 586-2945.

25
R. Neugebauer, B. Frazer, P. May-Ostendorp, C. Calwell, Ecos Consulting. Lowering the Cost of Play: Improving

the Energy Efficiency of Video Game Consoles. 2008. Natural Resources Defense Council: New York, NY.

26
Appliance Magazine. 56th Annual Industry Forecast. January, 2008. 65 (1): pp 38.
27
Saxman, Donald. Portable Battery Powered Products: Global Markets. 2008. BCC Research: Wellesley, MA.
Report No. FCB023D
28
American Pet Products Manufacturers Association. 2007-2008 national Pet Owners Survey. Quoted in: The Pet
Haven. Pet Ownership Statistics. July 18, 2007. (Last accessed September 13, 2010.) <http://the-pet
haven.blogspot.com/2007/07/pet-ownership-statistics.html> The September 13, 2010 version of this website is
available in Docket # EERE-2008-BT-STD-0005. For more information, contact Ms. Brenda Edwards at (202) 586
2945.
29
Graystone Creations. Frequently Asked Questions. (Last accessed July 15, 2010.)
<http://www.graystonecreations.com/definitions.htm> The July 15, 2010 version of this website is available in
Docket # EERE-2008-BT-STD-0005. For more information, contact Ms. Brenda Edwards at (202) 586-2945.
30
Southwest Garden Arts. FAQ. February 26, 2008. (Last accessed July, 14 2010.)
<http://www.arizonafountains.com/FAQ.htm> The July 14, 2010 version of this website is available in Docket #
EERE-2008-BT-STD-0005. For more information, contact Ms. Brenda Edwards at (202) 586-2945.
31
Appliance Magazine. 32nd Annual Portrait of the U.S. Appliance Industry. September, 2009. p. 34.
32
Teschke, D. A., P. Goldman, N. Levy, T. Goodman. Housewares MarketWatch. Fall 2007. International
Housewares Association: Rosemont, IL. Report No. 091707A/18M
33
Teschke, D. A., P. Goldman, J. Marshall-Bolton, T. Goodman. Housewares MarketWatch. Summer 2009.
International Housewares Association: Rosemont, IL.
34
Berner, Robert. Why P&Gs Smile is So Bright. Business Week. August 12, 2002. (Last accessed September 13,
2010.) <http://www.businessweek.com/magazine/content/02_32/b3795072.htm> The September 13, 2010 version of
this website is available in Docket # EERE-2008-BT-STD-0005. For more information, contact Ms. Brenda
Edwards at (202) 586-2945.
35
Toy Industry Association, Inc. State of the Industry. (Last accessed September 13, 2010.)
<http://www.toyassociation.org/AM/Template.cfm?Section=12_Rolling_Months> The September 13, 2010 version
of this website is available in Docket # EERE-2008-BT-STD-0005. For more information, contact Ms. Brenda
Edwards at (202) 586-2945.

3-78

36

International Market Solutions. Golf Car-Type Vehicles and the Emerging Market for Small, Task-Oriented

Vehicles in the United States. December 2007.

37
Toy Industry Association. State of the U.S. Toy Industry Rolling Report. July 2009.

38
United States Census Bureau. U.S. Population Projections. 2008.

39
New York Times. Goodman, David. "An Electric Boost for Bicyclists" February 1, 2010. (Last accessed

September 13, 2010.) <http://www.nytimes.com/2010/02/01/business/global/01ebike.html?pagewanted=all#> The

September 13, 2010 version of this website is available in Docket # EERE-2008-BT-STD-0005. For more

information, contact Ms. Brenda Edwards at (202) 586-2945.

40
Marketstrat, Inc. Total Powered Mobility Aids Market - United States (2004 - 2012). 2006.

41
Global Industry Analysts, Inc. Wheelchairs (Powered and Manual). March 2008.

42
U.S. Department of Energy. Draft Technical Report Energy Conservation Program for Consumer Products:

Battery Chargers and External Power Supplies. May 2009. Available at:

http://www1.eere.energy.gov/buildings/appliance_standards/residential/pdfs/bceps_draft_tech_report.pdf

43
US Environmental Protection Agency. Energy Star External Power Supply Qualified Products. September 2009.

44
Lawrence Berkeley National Laboratory. Climate Change Action Plan August 2009 Update. August, 2009.

45
Darnell Group, Inc. External AC-DC Power Supplies Second Edition. August 2005.

46
Title 10Energy, Chapter IIDepartment of Energy, Part 430Energy Conservation Program for Consumer

Products, Subpart BTest Procedures.

3-79

CHAPTER 4. SCREENING ANALYSIS


TABLE OF CONTENTS
4.1
4.2
4.3
4.4
4.5

INTRODUCTION ........................................................................................................... 4-1

SCREENED OUT EPS TECHNOLOGY OPTIONS ..................................................... 4-2

REMAINING EPS DESIGN OPTIONS ......................................................................... 4-2

SCREENED OUT BC TECHNOLOGY OPTIONS ....................................................... 4-2

REMAINING BC DESIGN OPTIONS........................................................................... 4-3

LIST OF TABLES

TABLE 4.1 EXTERNAL POWER SUPPLY DESIGN OPTIONS ............................................ 4-2

TABLE 4.2 SCREENED OUT BATTERY CHARGER TECHNOLOGY OPTIONS .............. 4-3

TABLE 4.3 BATTERY CHARGER DESIGN OPTIONS.......................................................... 4-3

4-i

CHAPTER 4. SCREENING ANALYSIS

4.1

INTRODUCTION

This chapter discusses the U.S. Department of Energys (DOEs) screening analysis of the
technology options identified for battery chargers (BC) and external power supplies (EPS). As
discussed in chapter 3 of the preliminary technical support document (TSD), DOE consults with
industry, technical experts, and other interested parties to develop a list of technology options for
consideration. The purpose of the screening analysis is to determine which options to consider
further and which to screen out.
Section 325(o)(2) of the Energy Policy and Conservation Act (EPCA) provides that any
new or revised standard must be designed to achieve the maximum improvement in energy
efficiency that is determined to be technologically feasible and economically justified. (42
U.S.C. 6295(o)(2)) In view of the EPCA requirements appendix A to subpart C of title 10, Code
of Federal Regulations (CFR), part 430 (10 CFR part 430), Procedures, Interpretations, and
Policies for Consideration of New or Revised Energy Conservation Standards for Consumer
Products (the Process Rule) sets forth procedures to guide DOE in its consideration and
promulgation of new or revised efficiency standards. These procedures elaborate on the statutory
criteria provided in 42 U.S.C. 6295(o) and, in part, eliminate problematic technologies early in
the process of prescribing or amending an energy efficiency standard. In particular, sections
4(b)(4) and 5(b) of the Process Rule provide guidance to DOE for determining which design
options are unsuitable for further consideration:
1. Technological feasibility. DOE will consider technologies incorporated in commercial
products or in working prototypes to be technologically feasible.
2. Practicability to manufacture, install, and service. If mass production and reliable
installation and servicing of a technology in commercial products could be achieved on
the scale necessary to serve the relevant market at the time the standard comes into effect,
then DOE will consider that technology practicable to manufacture, install, and service.
3. Adverse impacts on product utility or product availability. If DOE determines a
technology would have significant adverse impact on the utility of the product to
significant subgroups of consumers, or would result in the unavailability of any covered
product type with performance characteristics (including reliability), features, sizes,
capacities, and volumes that are substantially the same as products generally available in
the United States at the time, it will not consider this technology further.
4. Adverse impacts on health or safety. If DOE determines that a technology will have
significant adverse impacts on health or safety, it will not consider this technology
further.
Section 4.2 discusses the EPS technology options DOE screened out from further
consideration. Section 4.3 lists the remaining design options DOE considered in its analyses.

4-1

4.2

SCREENED OUT EPS TECHNOLOGY OPTIONS

DOE did not screen out any technology options, having considered the following four
factors: (1) technological feasibility; (2) practicability to manufacture, install, and service; (3)
adverse impacts on product utility to consumers; and (4) adverse impacts on health or safety.
4.3

REMAINING EPS DESIGN OPTIONS

DOE is considering the design options in Table 4.1 as viable means for improving EPS
efficiency. Chapter 3 provides a detailed description of these design options, which DOE will
consider in the engineering analysis (chapter 5).
Table 4.1 External Power Supply Design Options
Technology Option
Description
Improved Transformers
Use transformers with low losses.
Switched-Mode Power Supply
Use switched-mode power supplies instead of
linear power supplies.
Low-Power Integrated Circuits
Use integrated circuit controllers with minimal
power consumption.
Schottky Diodes and Synchronous
Use rectifiers with low losses.
Rectification
Low-Loss Transistors
Use transistors with low drain-to-source
resistance.
Resonant Switching
Use an algorithm to turn on the transformer
only when losses are minimal.
Resonant ("Lossless") Snubbers
Reuse energy sent to the snubber.
4.4

SCREENED OUT BC TECHNOLOGY OPTIONS

This section addresses the BC technologies that DOE screened out, having considered the
following factors: (1) technological feasibility; (2) practicability to manufacture, install, and
service; (3) adverse impacts on product utility to consumers; and (4) adverse impacts on health
or safety.
DOE examined all of the technology options presented in the technology assessment. The
table below lists out the options that DOE decided to screen out and which criterion that
technology option failed to meet:

4-2

Table 4.2 Screened Out Battery Charger Technology Options


Technology Option
Failed Screening Criterion
Lowering charging current or increasing
Adverse impacts on product utility to
voltage
consumers
Capacitive reactance
Adverse impacts on safety
Non-inductive chargers for toothbrush and
Adverse impacts on safety
other wet applications
4.5

REMAINING BC DESIGN OPTIONS

After screening out the aforementioned technology options in accordance with the policies
set forth in 10 CFR Part 430, Subpart C, Appendix A, (4)(a)(4) and 5(b), DOE is considering the
design options in Table 4.3 as viable means for improving battery charger efficiency. Chapter 3
provides a detailed description of these design options, which DOE will consider in the
engineering analysis (chapter 5).
Table 4.3 Battery Charger Design Options
Technology Option
Improved Cores
Termination

Slow charger

Fast charger

Description
Use transformer cores with low losses.
Limit power provided to fully-charged
batteries.
Elimination/Limitation of
Limit power provided to fully-charged
Maintenance Current
batteries.
Elimination of No-Battery Current Limit power provided drawn when no
battery is present.
Switched-Mode Power Supply
Use switched-mode power supplies instead
of linear power supplies.
Low-Power Integrated Circuits
Use integrated circuit controllers with
minimal power consumption.
Elimination of No-Battery Current Limit power provided drawn when no
battery is present.
Schottky Diodes and Synchronous Use rectifiers with low losses.
Rectification
Elimination of No-Battery Current Limit power provided drawn when no
battery is present.
Phase Control to Limit Input
Limit input power in lower-power modes.
Power

4-3

CHAPTER 5. ENGINEERING ANALYSIS

TABLE OF CONTENTS
CHAPTER 5. ENGINEERING ANALYSIS .............................................................................. 5-1

5.1
Introduction.......................................................................................................... 5-1

5.2
Engineering Analysis Methodology .................................................................... 5-1

5.3
Market Survey to Select Represenative Units ..................................................... 5-3

5.4
Class A External Power Supply Engineering Analysis ....................................... 5-8

5.4.1 Class A EPS Representative Product Classes and Representative Units. 5-8

5.4.2 Candidate Standard Levels for Representative Product Class A1 ......... 5-10

5.4.3 Class A EPS Testing Analysis ............................................................... 5-16

5.4.4 Class A EPS Teardowns to Estimate Manufacturer Production Cost.... 5-24

5.4.5 Class A EPS Manufacturer Interviews .................................................. 5-26

5.4.6 Class A EPS Scaling Relationships ....................................................... 5-36

5.4.7 Class A EPS Cost and Efficiency Relationships from Testing and

Teardowns.............................................................................................. 5-52

5.4.8 Class A EPS Cost and Efficiency Relationships from Manufacturer

Interviews............................................................................................... 5-54

5.4.9 Summary of Results for Class A EPSs .................................................. 5-57

5.4.10 Creating Class A EPS CSL Equations for All Product Classes Based On

the Representative Units ........................................................................ 5-63

5.5
Non-Class A External Power Supply Engineering Analysis ............................. 5-68

5.5.1 Non-Class A EPS Representative Product Classes and Representative

Units....................................................................................................... 5-68

5.5.2 Non-Class A EPS Candidate Standard Levels of Efficiency................. 5-71

5.5.3 Developing Cost-Efficiency Results for NCA EPSs ............................. 5-74

5.6
Battery Charger Engineering Analysis .............................................................. 5-75

5.6.1 BC Representative Product Classes ....................................................... 5-75

5.6.2 BC Representative Units........................................................................ 5-78

5.6.3 BC Candidate Standard Levels of Efficiency ........................................ 5-82

5.6.4 BC Teardown Analysis .......................................................................... 5-84

5.6.5 BC Manufacturer Interviews.................................................................. 5-85

5.6.6 BC Cost Model ...................................................................................... 5-92

5.6.7 BC Scaling from Representative Units to Product Classes ................... 5-94

5.6.8 Representative Product Class 1: Low Energy, Inductive....................... 5-96

5.6.9 Representative Product Class 2: Low Energy, Low Voltage............... 5-101

5.6.10 Representative Product Class 4: Low Energy, High Voltage .............. 5-113

5.6.11 Representative Product Class 5: Med. Energy, Low Voltage.............. 5-125

5.6.12 Representative Product Class 7: High Energy ..................................... 5-135

5.6.13 Representative Product Class 10: Low Energy, AC Output ................ 5-139

5.6.14 Extending the Analysis to Product Classes not Analyzed Directly ..... 5-148

5.6.14.4
Scaled Product Class 9: Low Energy, 12 V DC Input............. 5-153

5-i

LIST OF TABLES

TABLE 5.1 PROPOSED EPS REPRESENTATIVE PRODUCT CLASS AND SCALED

PRODUCT CLASSES..................................................................................................... 5-8

TABLE 5.2 EVALUATION OF POTENTIAL LOW-POWER EPSS AS REPRESENTATIVE

UNITS.............................................................................................................................. 5-9

TABLE 5.3 EXTERNAL POWER SUPPLY REPRESENTATIVE UNITS ........................... 5-10

TABLE 5.4 CANDIDATE STANDARD LEVELS OF EFFICIENCY FOR PRODUCT CLASS

A1................................................................................................................................... 5-10

TABLE 5.5 BASELINE (CSL0) EISA 2007 ACTIVE MODE EFFICIENCY FOR PRODUCT

CLASS A1 ..................................................................................................................... 5-12

TABLE 5.6 BASELINE (CSL0) EISA 2007 NO-LOAD POWER CONSUMPTION FOR

PRODUCT CLASS A1.................................................................................................. 5-12

TABLE 5.7 CSL1 ENERGY STAR 2.0 ACTIVE MODE EFFICIENCY FOR PRODUCT

CLASS A1 ..................................................................................................................... 5-13

TABLE 5.8 CSL1 ENERGY STAR 2.0 NO-LOAD MODE POWER CONSUMPTION FOR

PRODUCT CLASS A1.................................................................................................. 5-13

TABLE 5.9 CSL 2 INTERMEDIATE ACTIVE-MODE EFFICIENCY FOR PRODUCT CLASS

A1................................................................................................................................... 5-14

TABLE 5.10 CSL 2 INTERMEDIATE NO-LOAD MODE POWER CONSUMPTION FOR

PRODUCT CLASS A1.................................................................................................. 5-14

TABLE 5.11 CSL3 BEST IN MARKET ACTIVE MODE EFFICIENCY FOR PRODUCT

CLASS A1 ..................................................................................................................... 5-14

TABLE 5.12 CSL3 BEST IN MARKET NO-LOAD MODE POWER CONSUMPTION FOR

PRODUCT CLASS A1.................................................................................................. 5-15

TABLE 5.13 CSL4 MAX TECH ACTIVE MODE EFFICIENCY FOR PRODUCT CLASS A1

........................................................................................................................................ 5-16

TABLE 5.14 CSL4 MAX TECH NO-LOAD MODE POWER CONSUMPTION FOR

PRODUCT CLASS A1.................................................................................................. 5-16

TABLE 5.15 2.5-WATT UNITS USED TO CHARACTERIZE THE CSLS .......................... 5-20

TABLE 5.16 18-WATT UNITS USED TO CHARACTERIZE THE CSLS ........................... 5-21

TABLE 5.17 60-WATT UNITS USED TO CHARACTERIZE THE CSLS ........................... 5-22

TABLE 5.18 120-WATT UNITS USED TO CHARACTERIZE THE CSLS ......................... 5-24

TABLE 5.19 ISUPPLI TEARDOWN COST ESTIMATES..................................................... 5-26

TABLE 5.20 CURVES CHARACTERIZING THE RELATIONSHIP BETWEEN OUTPUT

VOLTAGE AND EFFICIENCY................................................................................... 5-45

TABLE 5.21 EFFECTS OF A 1.66M OUTPUT CORD ON EPS EFFICIENCY IN

PERCENTAGE POINTS .............................................................................................. 5-48

TABLE 5.22 AVERAGE OUTPUT CORD LENGTHS .......................................................... 5-49

TABLE 5.23 SCALING AND MARKUP STEPS PERFORMED ON THE TEST UNITS .... 5-53

TABLE 5.24 COST AND EFFICIENCY RELATIONSHIP FOR 2.5W EPS (TESTING AND

TEARDOWNS) ............................................................................................................. 5-53

TABLE 5.25 COST AND EFFICIENCY RELATIONSHIP FOR 18W EPS (TESTING AND

TEARDOWNS) ............................................................................................................. 5-54

TABLE 5.26 COST AND EFFICIENCY RELATIONSHIP FOR 60W EPS (TESTING AND

TEARDOWNS) ............................................................................................................. 5-54

5-ii

TABLE 5.27 COST AND EFFICIENCY RELATIONSHIP FOR 120W EPS (TESTING AND

TEARDOWNS) ............................................................................................................. 5-54

TABLE 5.28 SCALING AND MARKUP STEPS PERFORMED ON THE AGGREGATED

MANUFACTURER DATA .......................................................................................... 5-55

TABLE 5.29 COST AND EFFICIENCY RELATIONSHIP FOR 2.5W EPS

(MANUFACTURER INTERVIEWS) .......................................................................... 5-56

TABLE 5.30 COST AND EFFICIENCY RELATIONSHIP FOR 18W EPS

(MANUFACTURER INTERVIEWS) .......................................................................... 5-56

TABLE 5.31 COST AND EFFICIENCY RELATIONSHIP FOR 60W EPS

(MANUFACTURER INTERVIEWS) .......................................................................... 5-56

TABLE 5.32 COST AND EFFICIENCY RELATIONSHIP FOR 120W EPS

(MANUFACTURER INTERVIEWS) .......................................................................... 5-57

TABLE 5.33 ACTIVE-MODE EFFICIENCY CSL EQUATIONS FOR PRODUCT CLASSES

A1, A2, A3, AND A4. ................................................................................................... 5-65

TABLE 5.34 NO-LOAD POWER CSL EQUATIONS FOR PRODUCT CLASSES A1, A2, A3,

AND A4. ........................................................................................................................ 5-66

TABLE 5.35 THE LOW-VOLTAGE PRODUCT CLASSES (A2, A4) ACTIVE MODE

EFFICIENCY EQUATIONS......................................................................................... 5-67

TABLE 5.36 AC/AC PRODUCT CLASSES (A3, A4) MAXIMUM ALLOWABLE NO-LOAD

POWER EQUATIONS.................................................................................................. 5-68

TABLE 5.37 MULTIPLE VOLTAGE EPS REPRESENTATIVE PRODUCT CLASSES..... 5-69

TABLE 5.38 MULTIPLE-VOLTAGE EPS REPRESENTATIVE UNITS.............................. 5-69

TABLE 5.39 HIGH POWER EPS REPRESENTATIVE PRODUCT CLASSES ................... 5-69

TABLE 5.40 HIGH-POWER EPS REPRESENTATIVE UNITS ............................................ 5-69

TABLE 5.41 PROPOSED MEDICAL EPS REPRESENTATIVE PRODUCT CLASSES..... 5-70

TABLE 5.42 POSSIBLE MEDICAL EPS REPRESENTATIVE UNITS................................ 5-70

TABLE 5.43 PROPOSED MADB EPS REPRESENTATIVE PRODUCT CLASSES ........... 5-70

TABLE 5.44 POSSIBLE MADB EPS REPRESENTATIVE UNITS ...................................... 5-71

TABLE 5.45 MULTIPLE-VOLTAGE EPS CANDIDATE STANDARD LEVELS OF

EFFICIENCY ................................................................................................................ 5-71

TABLE 5.46 HIGH-POWER EPS CANDIDATE STANDARD LEVELS OF EFFICIENCY5-72

TABLE 5.47 MEDICAL EPS CANDIDATE STANDARD LEVELS OF EFFICIENCY ...... 5-72

TABLE 5.48 MADB EPS CANDIDATE STANDARD LEVELS OF EFFICIENCY............. 5-73

TABLE 5.49 REPRESENTATIVE PRODUCT CLASSES ..................................................... 5-78

TABLE 5.50 THE BC REPRESENTATIVE UNITS FOR EACH REPRESENTATIVE

PRODUCT CLASS. ...................................................................................................... 5-80

TABLE 5-51. ENERGY CONSUMPTION OF BATTERY CHARGER REPRESENTATIVE

UNITS AT EACH EFFICIENCY LEVEL.................................................................... 5-83

TABLE 5.52 WEEKLY USAGE ASSUMPTIONS FOR BATTERY CHARGER

REPRESENTATIVE UNITS ........................................................................................ 5-84

TABLE 5.53 AVERAGE MARKUP APPLIED TO THE BOM COSTS OBTAINED BOTH

THROUGH TEARDOWNS AS WELL AS MANUFACTURER INTERVIEWS TO

OBTAIN THE MSP....................................................................................................... 5-94

TABLE 5.54 CHARACTERISTICS OF TYPICAL UNITS FOR PRODUCT CLASS 1

DESCRIBED BY MANUFACTURERS DURING INTERVIEWS. ........................... 5-98

5-iii

TABLE 5.55 COMPARISON OF MSPS FOR PRODUCT CLASS 1S REPRESENTATIVE

UNIT OBTAINED FROM MANUFACTURER INTERVIEWS. NO TEARDOWNS

WERE PERFORMED FOR THIS REPRESENTATIVE UNIT................................. 5-101

TABLE 5.56 DETAILED TEST RESULTS FOR THE UNITS CORRESPONDING TO THE

CSLS FOR REPRESENTATIVE PRODUCT CLASS 2: LOW ENERGY, LOW

VOLTAGE................................................................................................................... 5-103

TABLE 5.57 CHARACTERISTICS OF TYPICAL UNITS FOR PRODUCT CLASS 2

DESCRIBED BY MANUFACTURERS DURING INTERVIEWS. ......................... 5-103

TABLE 5.58 COMPARISON OF MSPS FOR PRODUCT CLASS 2S REPRESENTATIVE

UNIT OBTAINED FROM MANUFACTURER INTERVIEWS VERSUS THOSE

DEVELOPED THROUGH TEARDOWNS. .............................................................. 5-112

TABLE 5.59 DETAILED TEST RESULTS FOR THE UNITS CORRESPONDING TO THE

CSLS FOR REPRESENTATIVE PRODUCT CLASS 4: LOW ENERGY, HIGH

VOLTAGE................................................................................................................... 5-115

TABLE 5.60 CHARACTERISTICS OF TYPICAL UNITS FOR PRODUCT CLASS 4

DESCRIBED BY MANUFACTURERS DURING INTERVIEWS. ......................... 5-116

TABLE 5.61 COMPARISON OF MSPS FOR PRODUCT CLASS 4S REPRESENTATIVE

UNIT OBTAINED FROM MANUFACTURER INTERVIEWS VERSUS THOSE

DEVELOPED THROUGH TEARDOWNS. .............................................................. 5-125

TABLE 5.62 DETAILED TEST RESULTS FOR THE UNITS CORRESPONDING TO THE

CSLS FOR REPRESENTATIVE PRODUCT CLASS 5: MEDIUM ENERGY, LOW

VOLTAGE................................................................................................................... 5-128

TABLE 5.63 CHARACTERISTICS OF TYPICAL UNITS FOR PRODUCT CLASS 5

DESCRIBED BY MANUFACTURERS DURING INTERVIEWS. ......................... 5-129

TABLE 5.64 COMPARISON OF MSPS FOR PRODUCT CLASS 5S REPRESENTATIVE

UNIT OBTAINED FROM MANUFACTURER INTERVIEWS VERSUS THOSE

DEVELOPED THROUGH TEARDOWNS. .............................................................. 5-135

TABLE 5.65 CHARACTERISTICS OF TYPICAL UNITS FOR PRODUCT CLASS 7

DESCRIBED BY MANUFACTURERS DURING INTERVIEWS. ......................... 5-136

TABLE 5.66 COMPARISON OF MSPS FOR PRODUCT CLASS 7S REPRESENTATIVE

UNIT OBTAINED FROM MANUFACTURER INTERVIEWS VERSUS THOSE

DEVELOPED THROUGH TEARDOWNS. .............................................................. 5-139

TABLE 5.67 DETAILED TEST RESULTS FOR THE UNITS CORRESPONDING TO THE

CSLS FOR PRODUCT CLASS 10: LOW-ENERGY AC OUTPUT. ........................ 5-141

TABLE 5.68 CHARACTERISTICS OF TYPICAL UNITS FOR PRODUCT CLASS 10

DESCRIBED BY MANUFACTURERS DURING INTERVIEWS. ......................... 5-142

TABLE 5.69 COMPARISON OF MSPS FOR PRODUCT CLASS 10S REPRESENTATIVE

UNIT OBTAINED FROM MANUFACTURER INTERVIEWS VERSUS THOSE

DEVELOPED THROUGH TEARDOWNS. .............................................................. 5-147

TABLE 5.70 BC SHIPMENTS-WEIGHTED AVERAGE USAGE PROFILES FOR

PRODUCT CLASSES 2, 3, AND 4 ............................................................................ 5-149

TABLE 5.71 CHARACTERISTICS OF TYPICAL UNITS SCALED FOR PRODUCT CLASS

3.................................................................................................................................... 5-150

TABLE 5.72 RESULTING COST AND EFFICIENCY RELATIONSHIP FOR BC PRODUCT

CLASS 3. ..................................................................................................................... 5-150

5-iv

TABLE 5.73 CHARACTERISTICS OF TYPICAL UNITS SCALED FOR PRODUCT CLASS

6.................................................................................................................................... 5-151

TABLE 5.74 RESULTING COST AND EFFICIENCY RELATIONSHIP FOR BC PRODUCT

CLASS 6. ..................................................................................................................... 5-151

TABLE 5.75 COST AND EFFICIENCY RELATIONSHIP FOR 2.5W EPS

(MANUFACTURER INTERVIEWS) ........................................................................ 5-152

TABLE 5.76 CHARACTERISTICS OF TYPICAL UNITS SCALED FOR PRODUCT CLASS

8.................................................................................................................................... 5-152

TABLE 5.77 RESULTING COST AND EFFICIENCY RELATIONSHIP FOR BC PRODUCT

CLASS 8. ..................................................................................................................... 5-152

TABLE 5.78 CHARACTERISTICS OF TYPICAL UNITS SCALED FOR PRODUCT CLASS

9.................................................................................................................................... 5-153

TABLE 5.79 RESULTING COST AND EFFICIENCY RELATIONSHIP FOR BC PRODUCT

CLASS 9. ..................................................................................................................... 5-153

LIST OF FIGURES
FIGURE 5.1. MARKET CHARACTERIZATION FOR POPULAR WALL ADAPTERS

(BUBBLE CHART)......................................................................................................... 5-5

FIGURE 5.2. MARKET CHARACTERIZATION FOR POPULAR WALL ADAPTERS (BAR

CHART)........................................................................................................................... 5-6

FIGURE 5.3 DISTRIBUTION OF BATTERY VOLTAGE AND ENERGY FOR HIGH


VOLUME APPLICATIONS IDENTIFIED IN THE MARKET SURVEY................... 5-7

FIGURE 5.4 CSL SCALED AVERAGE ACTIVE-MODE EFFICIENCY VS. OUTPUT

POWER.......................................................................................................................... 5-11

FIGURE 5.5 CSL NO-LOAD POWER VS. OUTPUT POWER.............................................. 5-12

FIGURE 5.6 SURVEY OF EPS EFFICIENCIES BY APPLICATION ................................... 5-17

FIGURE 5.7 NO-LOAD POWER VS. SCALED EFFICIENCY FOR 2.5-WATT UNITS..... 5-19

FIGURE 5.8 NO-LOAD POWER VS. SCALED EFFICIENCY FOR 18-WATT UNITS...... 5-20

FIGURE 5.9 NO-LOAD POWER VS. SCALED EFFICIENCY FOR 60-WATT UNITS...... 5-22

FIGURE 5.10 NO-LOAD POWER VS. SCALED EFFICIENCY FOR 120-WATT UNITS.. 5-23

FIGURE 5.11 FULL COST OF PRODUCT: BREAKDOWN OF PRODUCTION AND NON


PRODUCTION COSTS ................................................................................................ 5-25

FIGURE 5.12 SAMPLE EFFICIENCY AND MSP MANUFACTURER DATA FOR THE 2.5W

REPRESENTATIVE UNIT........................................................................................... 5-29

FIGURE 5.13 SAMPLE NO-LOAD POWER AND MSP MANUFACTURER DATA FOR THE

2.5W REPRESENTATIVE UNIT................................................................................. 5-30

FIGURE 5.14 ILLUSTRATION OF AGGREGATION METHOD IN TWO DIMENSIONS

FOR THE 2.5W REPRESENTATIVE UNIT ............................................................... 5-31

FIGURE 5.15 ILLUSTRATION OF AGGREGATION METHOD IN TWO DIMENSIONS

FOR THE 2.5W REPRESENTATIVE UNIT ............................................................... 5-32

FIGURE 5.16 SCALING STEPS TO NORMALIZE EFFICIENCIES AND COSTS ............. 5-38

FIGURE 5.17 SCALING AN EPS EFFICIENCY DATA POINT WITH NAMEPLATE

OUTPUT POWER......................................................................................................... 5-39

FIGURE 5.18 60-WATT EPS PRODUCT FAMILIES FOR OUTPUT VOLTAGE SCALING 5
42

5-v

FIGURE 5.19 HIGH-EFFICIENCY CURVE USED FOR OUTPUT VOLTAGE SCALING 5-42

FIGURE 5.20 PRODUCT FAMILIES USED TO CREATE LOW-EFFICIENCY CURVE FOR

OUTPUT VOLTAGE SCALING.................................................................................. 5-43

FIGURE 5.21 LOW-EFFICIENCY CURVE USED FOR OUTPUT VOLTAGE SCALING. 5-44

FIGURE 5.22 HIGH-EFFICIENCY AND LOW-EFFICIENCY CURVES FOR OUTPUT

VOLTAGE SCALING .................................................................................................. 5-44

FIGURE 5.23 2.5-WATT EPS PRODUCT FAMILIES FOR OUTPUT VOLTAGE SCALING 5
45

FIGURE 5.24 18-WATT EPS PRODUCT FAMILIES FOR OUTPUT VOLTAGE SCALING 5
46

FIGURE 5.25 60-WATT EPS PRODUCT FAMILIES FOR OUTPUT VOLTAGE SCALING 5
46

FIGURE 5.26 120-WATT EPS PRODUCT FAMILIES FOR OUTPUT VOLTAGE SCALING

........................................................................................................................................ 5-47

FIGURE 5.27 CORD LENGTHS OF TESTED EPS UNITS ................................................... 5-49

FIGURE 5.28 THE FULL MARKUP CHAIN, INCLUDING THE STEPS FROM BOM TO

MPC TO MSP................................................................................................................ 5-52

FIGURE 5.29 2.5W MSP VS. EFFICIENCY CURVES FOR MANUFACTURER AND

TESTING AND TEARDOWN DATA ......................................................................... 5-59

FIGURE 5.30 2.5W MSP VS. NO-LOAD POWER CURVES FOR MANUFACTURER AND

TESTING AND TEARDOWN DATA ......................................................................... 5-59

FIGURE 5.31 18W MSP VS. EFFICIENCY CURVES FOR MANUFACTURER AND

TESTING AND TEARDOWN DATA ......................................................................... 5-60

FIGURE 5.32 18W MSP VS. NO-LOAD POWER CURVES FOR MANUFACTURER AND

TESTING AND TEARDOWN DATA ......................................................................... 5-60

FIGURE 5.33 60W MSP VS. EFFICIENCY CURVES FOR MANUFACTURER AND

TESTING AND TEARDOWN DATA ......................................................................... 5-61

FIGURE 5.34 60W MSP VS. NO-LOAD POWER CURVES FOR MANUFACTURER AND

TESTING AND TEARDOWN DATA ......................................................................... 5-62

FIGURE 5.35 120W MSP VS. EFFICIENCY CURVES FOR MANUFACTURER AND

TESTING AND TEARDOWN DATA ......................................................................... 5-63

FIGURE 5.36 120W MSP VS. NO-LOAD POWER CURVES FOR MANUFACTURER AND

TESTING AND TEARDOWN DATA ......................................................................... 5-63

FIGURE 5.37. DISTRIBUTION OF ANNUAL SHIPMENTS BY PRODUCT CLASS. ....... 5-77

FIGURE 5.38. DISTRIBUTION OF ANNUAL ENERGY CONSUMPTION BY PRODUCT

CLASS. .......................................................................................................................... 5-77

FIGURE 5.39 THE BC REPRESENTATIVE UNITS SUPERIMPOSED OVER THE BC

MARKET SURVEY RESULTS. .................................................................................. 5-81

FIGURE 5.40 REPRESENTATIVE UNITS AND UNITS DESCRIBED BY

MANUFACTURERS. ................................................................................................... 5-86

FIGURE 5.41 ILLUSTRATION OF THE REGRESSION ANALYSIS OF MANUFACTURER

COST-EFFICIENCY DATA......................................................................................... 5-89

FIGURE 5.42 ILLUSTRATION OF TRANSLATION OF CURVES TO OBTAIN

INCREMENTAL COSTS FROM THE BASELINE .................................................... 5-90

FIGURE 5.43 AGGREGATION OF TRANSLATED MANUFACTURER CURVES........... 5-90

5-vi

FIGURE 5.44 TRANSLATION OF CURVES TO OBTAIN ABSOLUTE COSTS FROM

TEARDOWN BASELINE ............................................................................................ 5-91

FIGURE 5.45 THE FOUR WAYS THAT BC AND END-USE PRODUCT MANUFACTURE

CAN BE DIVIDED BETWEEN THE BC SUPPLIER OR ORIGINAL DEVICE

MANUFACTURER (ODM) AND THE ORIGINAL EQUIPMENT MANUFACTURER

(OEM). ........................................................................................................................... 5-93

FIGURE 5.46 24-HOUR EFFICICIENCY STANDARD PROPOSED BY PG&E, WHICH

COULD BE USED TO SCALE THE ANALYSIS OF THE REPRESENTATIVE UNIT

TO BCS WITH DIFFERING BATTERY ENERGIES................................................. 5-96

FIGURE 5.47 TEST RESULTS FOR REPRESENTATIVE PRODUCT CLASS 1: LOW

ENERGY, INDUCTIVE................................................................................................ 5-97

FIGURE 5.48 ILLUSTRATION OF THE MSP FOR THIS REPRESENTATIVE UNIT AT

EACH CSL. ................................................................................................................. 5-100

FIGURE 5.49 TEST RESULTS FOR REPRESENTATIVE PRODUCT CLASS 2: LOW

ENERGY, LOW VOLTAGE. ..................................................................................... 5-102

FIGURE 5.50 INPUT POWER WAVEFORM DURING CHARGE AND MAINTENANCE

MODES FOR THE BASELINE UNIT. THE FULL 24-HOUR PLOT (A) SHOWS

SOME INCREASED ACTIVITY IN MAINTENANCE, POSSIBLY ASSOCIATED

WITH PHONE OPERATION, WHILE THE SIX-HOUR FRAGMENT (B) SHOWS

THE EXPONENTIAL CHARACTERISTIC TYPICAL OF SLOW CHARGERS. .. 5-106

FIGURE 5.51. THE TOP AND BOTTOM OF THE SLIGHTLY LESS-EFFICIENT BEST-IN
MARKET CAMERA CHARGER. NOTE THE LACK OF ANY INTEGRATED

CIRCUIT CONTROLLER. ......................................................................................... 5-108

FIGURE 5.52 PCBS OF DIGITAL CAMERA CHARGERS MEETING CSL 1. NOTE THE

HIGHLIGHTED MICROCONTROLLER IN A BLACK 14-PIN SURFACE-MOUNT

PACKAGE IN THE LEFT PHOTOGRAPH. SIMILAR FUNCTIONALITY IS

PROVIDED IN THE OTHER CHARGER BY THE DEDICATED CHARGER IC

HIGHLIGHTED AT THE RIGHT.............................................................................. 5-109

FIGURE 5.53FRAGMENT OF THE INPUT POWER WAVEFORM DURING CHARGE AND

MAINTENANCE MODES FOR THE LESS EFFICIENT CSL2 DIGITAL CAMERA

CHARGER. ................................................................................................................. 5-110

FIGURE 5.54 FRAGMENT OF THE INPUT POWER WAVEFORM DURING CHARGE

AND MAINTENANCE MODES FOR THE MORE EFFICIENT CSL2 DIGITAL

CAMERA CHARGER. ............................................................................................... 5-111

FIGURE 5.55 MSPS FOR EACH CSL FOR REPRESENTATIVE UNIT 2. ........................ 5-112

FIGURE 5.56 RESULTS OF UNIVERSAL BATTERY CHARGER TEST RESULTS

PERFORMED BY ECOS CONSULTING WITH BATTERIES OF VARYING

ENERGIES. ALL TEST RESULTS FOR A GIVEN CHARGER ARE PERFORMED

WITH BATTERIES WITH THE SAME VOLTAGE. ............................................... 5-113

FIGURE 5.57 ILLUSTRATION OF THE LIMITED OVERLAP IN BC CHARACTERISTICS

BETWEEN NOTEBOOK COMPUTERS AND MOTOR-OPERATED APPLIANCES

IN THE LOW-ENERGY HIGH-VOLTAGE PRODUCT CLASS (BASED ON

ADVERTISED BATTERY VOLTAGE AND ENERGY)......................................... 5-114

FIGURE 5.58 TEST RESULTS FOR REPRESENTATIVE PRODUCT CLASS 4: LOW

ENERGY, HIGH VOLTAGE ..................................................................................... 5-115

5-vii

FIGURE 5.59 PHOTOGRAPHS OF THE TOP AND BOTTOM OF THE MOTHERBOARD

OF THE NOTEBOOK COMPUTER, WITH POWER MANAGEMENT AND

BATTERY CHARGING COMPONENTS HIGHLIGHTED..................................... 5-116

FIGURE 5.60 THE INPUT POWER OF THE BASELINE UNIT DURING CHARGING AND

MAINTENANCE MODES. ........................................................................................ 5-119

FIGURE 5.61FRAGMENT OF THE INPUT POWER WAVEFORM DURING CHARGE AND

MAINTENANCE MODES FOR THE CSL 1 UNIT.................................................. 5-120

FIGURE 5.62 THE EFFICIENCY RELATED MANUFACTURING COST OF THE FOUR

COMMERCIALLY ANALYZED BCS IN THIS PRODUCT CLASS. .................... 5-125

FIGURE 5.63 TEST RESULTS FOR REPRESENTATIVE PRODUCT CLASS 5: MEDIUM

ENERGY, LOW VOLTAGE. ..................................................................................... 5-127

FIGURE 5.64 PHOTOGRAPHS OF THE BASELINE MARINE CHARGER, SHOWING (A)

THE LINE-FREQUENCY TRANSFORMER AND (B) HEATSINK-MOUNTED SCR.

NOTE THE DIFFERENCE IN SIZE AND DESIGN COMPARED TO THE LOW


ENERGY BCS............................................................................................................. 5-130

FIGURE 5.65 FRAGMENT OF THE INPUT POWER WAVEFORM DURING CHARGE

AND MAINTENANCE MODES FOR THE BASELINE UNIT. .............................. 5-131

FIGURE 5.66 PHOTOGRAPHS OF THE IMPROVED MARINE CHARGER, SHOWING (A)

THE LINE-FREQUENCY TRANSFORMER AND (B) COMPLEX CHARGE


CONTROL PCB. ......................................................................................................... 5-132

FIGURE 5.67FRAGMENT OF THE INPUT POWER WAVEFORM DURING CHARGE AND

MAINTENANCE MODES FOR THE CSL 1 UNIT.................................................. 5-133

FIGURE 5.68 COST-EFFICIENCY RESULTS AT EACH CSL FOR REPRESENTATIVE

UNIT 5. ........................................................................................................................ 5-135

FIGURE 5.69 MANUFACTURER COST EFFICIENCY RELATIONSHIP FOR

REPRESENTATIVE UNIT 7...................................................................................... 5-138

FIGURE 5.70 TEST RESULTS FOR THE LOW-ENERGY, AC-OUTPUT

REPRESENTATIVE PRODUCT CLASS. ................................................................. 5-140

FIGURE 5.71 INPUT POWER WAVEFORM DURING CHARGE AND MAINTENANCE

MODES FOR THE CSL 1 UNIT. ............................................................................... 5-143

FIGURE 5.72 PHOTOGRAPH OF THE PCB OF THE BASELINE UNIT FOR THIS

PRODUCT CLASS. THE BC CIRCUITRY IS MAINLY TO THE LEFT................ 5-144

FIGURE 5.73 INPUT POWER WAVEFORM DURING CHARGE AND MAINTENANCE

MODES FOR THE CSL 1 UNIT. ............................................................................... 5-145

FIGURE 5.74 TEARDOWN AND INTERVIEW MSPS AT EACH CSL FOR

REPRESENTATIVE UNIT 10.................................................................................... 5-147

5-viii

CHAPTER 5. ENGINEERING ANALYSIS

5.1

INTRODUCTION

The U.S. Department of Energy (DOE) performed an engineering analysis to establish


the relationship between the manufacturer selling price (MSP) and the energy efficiency of
battery chargers (BCs) and external power supplies (EPSs). The relationship between the MSP
and energy efficiency, or the cost-efficiency relationship, serves as the basis for cost-benefit
calculations in terms of individual customers, manufacturers, and the Nation. This section
provides an overview of the engineering analysis, discusses the representative product classes,
establishes baseline unit specifications for those product classes, discusses incremental efficiency
levels, discusses the analysis and results for the representative product classes, and establishes a
scaling methodology to those product classes not analyzed directly.
The primary inputs of the engineering analysis are cost and efficiency data derived from
1) test data and teardown analysis, and 2) manufacturer interviews. Additional inputs include
design options from the screening analysis (preliminary technical support document (TSD)
chapter 4). The primary output of the engineering analysis is a set of cost-efficiency curves. In a
subsequent life-cycle cost analysis (preliminary TSD chapter 8), DOE used the cost-efficiency
curves to determine customer prices for each of the products analyzed in the engineering analysis
by applying the appropriate distribution channel markups.
5.2

ENGINEERING ANALYSIS METHODOLOGY

DOE structured its engineering analysis using two methodologies: (1) the testing and teardown
approach, which involves testing commercially available products and tearing down the products
to determine bottom-up manufacturing costs based on a detailed bill of materials and (2) the
efficiency-level approach, which involves interviewing manufacturers to determine the relative
costs of achieving increases in efficiency, without regard to the particular design options used to
achieve such increases. The following summarizes the general steps taken throughout the
engineering analysis:
Market Survey: DOE surveyed applications that use BCs and EPSs to determine the most
popular units in the market. DOE focused its analysis on these popular units.
Representative Product Classes and Representative Units: DOE reviewed covered BCs and
EPSs and their associated product classes. DOE selected certain classes and units as
representative and concentrated its analytical effort on these because they represent a
significant majority of units and because analysis on these units and classes can be extended to
all units and classes. For those product classes that are not analyzed directly, DOE extrapolates
the analysis from representative product classes.
Baseline Efficiency Level: For all representative units, DOE establishes baseline efficiency
levels, which serve as reference points against which DOE measures changes resulting from

5-1

potential amended energy conservation standards. To determine energy savings and changes in
price, DOE compares each higher energy-efficiency level with the baseline efficiency level. For
EPSs the baseline efficiency level is determined by Federal energy conservation standards set by
EISA 2007. Because of the wide variety of designs that manufacturers use to meet EISA 2007
standards, DOE does not describe specific baseline units with particular designs. There are no
existing federal standards for BCs, so the baseline BC reflects the lowest-efficiency units in the
market.
Unit Testing: DOE purchased and measured the efficiency of BC and EPS units to characterize
the full range of efficiencies in the market and the ranges of efficiencies of the representative
units.
Candidate Standard Levels. After identifying baseline efficiency levels, DOE developed
candidate standard levels based on: (1) voluntary efficiency specifications, (2) commercially
available high-efficiency units determined by testing units, (3) intermediate points in the market
and (4) the maximum technologically feasible (max-tech) efficiency level determined by
interviewing manufacturers of BCEPS, integrated-circuit controllers for BCEPS, and
applications that use BCEPS. The max-tech level was independently verified by subject matter
experts.
Unit Teardowns: DOE selected certain test units to characterize the costs for the representative
units to meet the CSLs. To determine costs, DOE subcontracted iSuppli Corporation, an industry
expert in costs of consumer electronics, to perform teardowns.
Manufacturer Interviews: DOEs contractors interviewed manufacturers of BCEPS,
manufacturers of integrated circuit (IC) controllers for BCEPS, and manufacturers of products
with applications that use BCEPS. During these interviews DOE obtained confidential design
cost and efficiency information which DOE aggregated to derive manufacturer-based
relationships between cost and efficiency for BCEPS.
Price Analysis. The costs output from the engineering analysis is at the point in the product value
chain where the BCEPS manufacturer sells its product to the application manufacturer, termed
the manufacturer selling price (MSP). However, in some cases DOE obtained costs for BCs and
EPSs earlier in the production process such as the (1) bill of materials (BOM), which describes
the products components in detail, including all manufacturing steps required to make and/or
assemble each part, or the (2) manufacturer production costs (MPCs), which is the cost of the BC
or EPS after it leaves the factory. By applying manufacturer markups to the BOMs and MPCs,
DOE calculated the MSPs used in the final cost-efficiency curves.
The sections that follow discuss how DOE applied this methodology to create the engineering
analysis.

5-2

5.3

MARKET SURVEY TO SELECT REPRESENATIVE UNITS

DOE began the development of the preliminary analysis with a market survey in the
summer of 2009 as the basis for selecting representative units for both BCs and Class A EPSs.
At the time of the survey non-Class A EPSs were part of a separate determination analysis and
therefore were addressed separately in that rulemaking. The goal of the survey for BCs and
Class A EPSs was to determine market segments from which to select the representative units.
To best capture the BC and Class A EPS markets, DOE focused on the most popular applications
that use BCs and Class A EPSs. The survey consisted of the following steps:
1. Identify the types of applications that use BCs and Class A EPSs
2. Estimate the annual shipments and the energy consumption for those applications
3. Select applications to focus on based on significant shipments and/or energy consumption
4. Visit websites and retail outlets to survey product characteristics of popular models
5. Combine the results into BC and Class A EPS market profiles
6. Select representative units based on the market profiles
As discussed in Chapter 3 of this preliminary TSD, to date DOE identified 83 applications
that use BCs and/or EPSs. When DOE conducted a market survey in the summer of 2009 it had
identified 51 applications. Of those applications, DOE had obtained estimated annual shipments
for 43. Among those applications, DOE prioritized EPS-focused research on 13 application
types that were expected to have large shipments and/or energy consumption.
1. Computer Speakers
2. Cordless Phones/Answering Devices
3. Digital Photo Frames
4. E-Books
5. External Hard Drives
6. Inkjet MFDs and Printers
7. LAN Equipment
8. Mobile Phones
9. MP3 Speaker Docks
10. Notebook Computers
11. Portable DVD Players
12. USB Wall Adapters
13. VoIP Adapters
DOE conducted similar model counts to assess models that would most impact the BC
analysis.
1. Camcorders
2. Cordless Phones/Answering Devices
3. Digital Cameras
4. E-Books
5. Electric Bicycles
6. Golf Carts
7. Handheld Vacuum Cleaners
8. Handheld Video Games
9. Hedge and Lawn Tools

5-3

10. Marine Chargers


11. Mobile Phones
12. Notebook Computers
13. Power Tools
14. Robotic Vacuum Cleaners
15. Uninterruptible Power Supplies
16. Universal Battery Chargers
For the selected applications, DOE identified the most popular product models,
considered to be those on the best-seller lists of several popular online retailers (e.g.,
Amazon.com, BestBuy.com, etc.). In total, DOE identified 366 application models among the
13 EPS-focused application types. For each model, DOE noted whether the model included a
wall adapter, a battery, both, or neither. Among the 366 models, 281 were identified as having
wall adapters, and DOE was able to discern the expected nameplate output power for 230 of the
models from 1.2 watts for a cordless phone to 120 watts for a notebook computer. In cases
where the BC or EPS information was not readily available, DOE consulted publicly available
manufacturer information and noted the characteristics of the product.
Subsequently, for EPSs, DOE combined the information it gathered on application
shipments, model count, and model power rating to characterize popular wall adapters by
nameplate output power, which DOE used as an approximation of the EPS market. Figure 5.1
and Figure 5.2 show the results of the market characterization in two formats: bubble chart and
bar chart, respectively. The size of the bubbles in Figure 5.1 gives a relative measure of how
many models DOE counted for each application at each nameplate output power rating. The
height of a bubble corresponds to the total shipments of the application. For example, mobile
phones have the highest shipments at approximately 143M per year.

5-4

1,000,000,000

Annual Shipments

100,000,000

10,000,000

1,000,000

100,000

12

16

20

24

28

32

36

40

44

48

52

56

60

64

Nameplate Output Power [W]


Computer speakers

Cordless Phones/Ans wering Devices

Digital Photo Frames

E-Books

External hard drives

Inkjet MFDs and printers

LAN Equipment

Mobile phones

MP3 Speaker Docks

Notebook computers

Portable DVD players

VoIP Adapters

Figure 5.1. Market Characterization for Popular Wall Adapters (bubble chart)
In Figure 5.2 application shipments have been apportioned relative to model count; the
height of each bar indicates the cumulative EPS shipments at a particular nameplate output
power. For example, the 143M shipments for mobile phones are divided between wall adapters
with nameplate output power ratings of approximately 3 watts and 5 watts, 16M and 127M
respectively. The bar on the x-axis corresponding to 3-watt EPSs consists of the 16M mobile
phone EPSs and 7M cordless phone EPSs. Thus the total estimated shipments for 3-watt EPSs
are 23M units per year. These volumes were a major factor in the process of selecting the
representative units, which is detailed in 5.4.1.

5-5

VoIP Adapters

150,000,000

Portable DVD players

135,000,000

Notebook computers
120,000,000

Annual Shipments

MP3 Speaker Docks

105,000,000

Mobile phones

90,000,000

LAN Equipment

75,000,000

Inkjet MFDs and printers


External hard drives

60,000,000

E-Books

45,000,000

Digital Photo Frames

30,000,000

Cordless Phones/Answering Devices

15,000,000

Computer speakers

0 0 1 1 1 2 3 4 5 6 7 8 9 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64 66 68 70

Nameplate Output Power [W]

Figure 5.2. Market Characterization for Popular Wall Adapters (bar chart)
While the EPS products were grouped by nameplate output power, the BC product
classesdescribed in chapter 3are defined by battery voltage (in volts) and energy (in watthours). Therefore, DOE grouped the models evaluated during the market survey by battery
voltage and energy. A comparison of the resultant market distribution to the BC product classes
can be seen in Figure 5.3. This market distribution was subsequently used to select the BC
representative product classes and units, as detailed in sections 5.6.1 and 5.6.2, respectively.

5-6

Figure 5.3 Distribution of battery voltage and energy for high-volume applications identified in the market survey

5-7

5.4

CLASS A EXTERNAL POWER SUPPLY ENGINEERING ANALYSIS

In this section, DOE presents a full engineering analysis including cost-efficiency curves
for Class A EPSs. In section 5.5, DOE presents a preliminary discussion of the engineering
analysis for non-Class A EPSs that includes a description of representative products classes,
representative units, and candidate standard levels of efficiency. For non-Class A EPSs, a more
detailed engineering analysis including product testing and teardown results, and manufacturerderived cost-efficiency curves will be derived for the NOPR analysis.
5.4.1

Class A EPS Representative Product Classes and Representative Units

DOE elected to focus the engineering analysis on the AC-DC basic-voltage output
product class A1 as the representative product class among the four product classes shown in
Table 5.1 because the majority of units in the market are in that product class. The other three
product classes were not explicitly analyzed in an engineering analysis using physical test units
or manufacturer data; rather, the analysis for product class A1 was extended to the other three
product classes using scaling relationships (discussed further in section 5.4.6 and section 5.4.10).
Product class A1 includes EPSs for low-power products such as cordless phones and cellular
phones, mid-power products such as modems and routers, and higher-power products such as
notebook computers.
Table 5.1 Proposed EPS Representative Product Class and Scaled Product Classes
Basic Voltage Output
Low Voltage Output *
AC-DC Conversion A1, Representative Product Class
A2, Scaled
AC-AC Conversion
A3, Scaled
A4, Scaled
* Low voltage output EPSs have nameplate output voltage less than six volts and nameplate output current greater
than or equal to 550 milliamps. All other EPSs are basic voltage output.

Within the representative product class, DOE chose representative units based on the
following criteria:
Select units only within the representative product class A1
Select a sufficient amount of nameplate output power points to curve fit the CSLs
Focus on popular areas in the market
Ensure that there are many units to test
DOE chose to have multiple representative units for the EPS engineering analysis since
EPS product classes span a wide range of output power, and efficiency is strongly affected by
output power. Since DOE elected to express the CSLs as continuous functions of nameplate
output power it was necessary to select several representative units to characterize each CSL. To
accomplish that goal, DOE chose to characterize four nameplate output power regions: low
power (1.2 watts to 10 watts), mid power (15 watts to 30 watts), high power (30 watts to 90
watts), and maximum power of 120 watts. The representative units are focused on lower powers
because there lies the greatest diversity of EPS efficiency. Above 50 watts, there is much less
variation in EPS efficiency, hence there is less need for characterization.
5-8

After selecting the number of representative units, DOE selected each units nameplate
output power and nameplate output voltage the two characteristics that most influence EPS
efficiency. The combination of power and voltage for each representative unit needed to result
in the EPS being in the representative product class A1, per DOEs decision to analyze only that
product class. Further, by ensuring that all units were from the same product class, DOE could
scale results across all nameplate output powers within the representative product class during
the later stages of the analysis.
When defining the representative unit values, DOE considered specifying the same
output voltage for all representative units, but instead specified different output voltages for each
representative unit. DOE took this approach because there is a trend in the market for EPSs with
higher nameplate output power to have higher nameplate output power. For instance, low-power
cell phone chargers tend to have nameplate output voltage at 5 volts whereas higher-power
notebook computers tend to have nameplate output voltage at 19 volts. Thus, there is not a single
output voltage for all representative units that would accurately reflect the market. Further,
because the representative units have the different wattages and voltages, DOE ensured that the
CSLs are consistent with the market, since the representative units characterize the CSLs.
DOEs next step in selecting representative units was to focus on popular units in the
market, which was straightforward for most representative units, except the low-power
representative unit. Figure 5.2 shows a peak at 5 watts in the EPS market. Although DOE
considered a 5-watt representative unit because of its prevalence, DOE instead selected a 2.5
watt representative unit because it belonged to the representative product class A1 whereas the 5
watt EPS was in product class A2 (shown in Table 5.2). Specifically, EPSs with both nameplate
output voltage less than 6 volts and nameplate output current greater than or equal to 0.55
amperes are considered to be low-voltage EPSs that are in product class A2 (if they output DC
power) or product class A4 (if they output AC power). All of the 5-watt EPSs had nameplate
output voltage of 5 volts, nameplate output current of 1 ampere, and DC output power. Thus all
5-watt EPSs were in product class A2 and not viable for selection as representative units in the
representative product class A1. In contrast, 2.5-watt EPSs had nameplate output voltage of 5
watts and nameplate output current of 0.5 amperes. Thus the 2.5-watt EPSs are in product class
A1 and eligible as representative units. As noted in the executive summary, DOE is seeking
stakeholder comment on this approach in general and on whether to perform a detailed analysis
for 5-watt, 5-volt EPSs in particular.
Table 5.2 Evaluation of Potential Low-Power EPSs as Representative Units
Power
Voltage
Current
[W]
[V]
[A]
Low-voltage criteria
<6V
0.55 A
Valid representative unit
2.5 W
5V
0.50 A
Invalid representative unit
5W
5V
1.00 A

Product
Class
A2 and A4
A1
A2

The last criterion that DOE considered in selecting EPS representative units was unit
availability. This requirement was not trivial because, although there are many applications sold
with EPSs, a significant amount of those applications do not publish the nameplate output power

5-9

and voltage of their associated EPSs. Therefore, DOE targeted its analysis on EPSs that were
clearly available for purchase at specific output powers. Based on all of the criteria presented,
DOE selected four representative units for EPSs, listed in Table 5.3.
Table 5.3 External Power Supply Representative Units
Representative Nameplate Output
Nameplate Output
Unit
Power
Voltage
[watts]
[volts]
1
2
3
4

2.5
18
60
120

5
12
15
19

Example
Application

Mobile phone
Modem
Laptop Computer
Laptop Computer

Output
Cord
Length*
[m]
1.66
1.66
1.66
1.66

*The standard cord length assumed for all representative units was 1.66m. Cord length did not influence the
selection of units; it was only used to scale efficiency (see Section 5.4.6.3 for details) and cost (see Section 5.4.6.5
for details) data.

See Section 5.4.3 for a detailed discussion of all the EPSs tested, including those at the
representative units.
5.4.2

Candidate Standard Levels for Representative Product Class A1

DOE determined the CSLs for the AC-DC basic-voltage representative product class A1
based on existing standard levels, products available in the market, and information from
manufacturers, in the manner shown in Table 5.4.
Table 5.4 Candidate Standard Levels of Efficiency for Product Class A1
Number Reference
Basis
CSL0
EISA 2007
EISA 2007 equations for efficiency and no-load power
CSL1
Energy Star 2.0 Energy Star 2.0 equations for efficiency and no-load power
CSL2
Intermediate
Curve fit to manufacture data points
CSL3
Best in Market Curve fit to test unit data points
CSL4
Max Tech
Curve fit to manufacture data points
The CSL equations, as well as the representative unit test and manufacturer data are
illustrated in Figure 5.4 for efficiency, and in Figure 5.5 for no-load power. A discussion of the
exact equations and reasoning underlying each of the four CSLs illustrated in Figure 5.4 and
Figure 5.5, follows, starting with CSL0.
Figure 5.4 shows CSL equations that are higher in efficiency with higher CSLs, for a
given nameplate output power. This represents increasingly stringent standards with higher
CSLs. For each of the representative units and CSLs there is a corresponding aggregated
manufacturer data point and testing and teardown data point, except for CSL4, which has no test
data because it is the max-tech level, and CSL2, which has no test data because it was chosen to
be an intermediate level between the Energy Star 2.0 and the best-in-market levels. As shown in
Figure 5.4, the CSLs are fit closely to the data (see Section 5.4.10 for details on curve-fitting),

5-10

with the added constraint that they never go above a manufacturer or test data point in efficiency,
for a given nameplate output power.

Scaled Average Active-Mode


Efficiency vs. Output Power
Scaled Average Active-Mode Efficiency [%]

95%
90%

#951

#838
#941

#834

#118

#999

85%
80%

#854

Test Data
Manufacturer Data

EISA

#949

CSL1

75%

#996

CSL2
CSL3

70%
#935

CSL4

65%
#876

60%
55%
0

20

40

60

80

100

120

Output Power [W]

Figure 5.4 CSL Scaled Average Active-Mode Efficiency vs. Output Power
Figure 5.5 shows CSL equations that are lower or equal in no-load power with higher
CSLs, for a given nameplate output power. This represents increasingly stringent standards with
higher CSLs. For each of the representative units and CSLs there is a corresponding aggregated
manufacturer data point and testing and teardown data point, except for CSL4, which has no test
data because it is the max-tech level, and CSL2, which has no test data because it was chosen to
be an intermediate level between the Energy Star 2.0 and the best-in-market levels. As shown in
Figure 5.4, the CSLs are fit closely to the data (see Section 5.4.10 for details on curve-fitting),
with the added constraint that they never go below a manufacturer or test data point in no-load
power, for a given nameplate output power.

5-11

No-Load Power vs. Output Power

0.5
Test Data

Manufacturer Data
No-Load Power [W]

0.4

#876

0.3

CSL0
#949

CSL1

#118

CSL2
#854
#999

0.2

#951

CSL3
CSL4

#935
#834

#941

0.1

996
#838

0
0

20

40

60

80

100

120

Output Power [W]

Figure 5.5 CSL No-Load Power vs. Output Power


Currently, EPCA, as amended by EISA 2007, determines EPS minimum efficiency and
maximum no-load power consumption as a function of nameplate output power (Pout), shown in
Table 5.5 and Table 5.6 , respectively. For this analysis, the baseline efficiency level, CSL0, for
each representative unit is the EISA level at the corresponding nameplate output power.
Table 5.5 Baseline (CSL0) EISA 2007 Active Mode Efficiency for Product Class A1
Nameplate Output Power
Active Mode Required Efficiency
(decimal equivalent of a percentage)
<1 Watt
0.5 * Pout
1 to not more than 51 Watts

0.09 * ln (Pout) + 0.5

>51 Watts

0.85

Table 5.6 Baseline (CSL0) EISA 2007 No-Load Power Consumption for Product Class A1
Nameplate Output Power Maximum Power Consumption
Not more than 250 Watts
0.5
Energy Star 2.0 is another level in the market, although it is a voluntary specification
rather than a mandatory standard. Similar to EISA, the Energy Star 2.0 level is a function of

5-12

nameplate output power, shown in Table 5.7 and Table 5.8. As explained in chapter 3, Energy
Star 2.0 has four product classes distinguished by the type of output power (AC/DC or AC/AC)
and the output voltage (basic or low). DOE developed its product class structure in the same
manner as Energy Star, hence DOE chose CSL1 to be the same as the Energy Star 2.0 level for
AC-DC basic output voltage EPSs, at the corresponding nameplate output power.
Table 5.7 CSL1 Energy Star 2.0 Active Mode Efficiency for Product Class A1
Nameplate Output Power Minimum Average Efficiency in Active Mode
(expressed as a decimal)
0 to 1 watt
0.480 * Pout + 0.140
> 1 to 49 watts

0.0626 * ln (Pout) + 0.622

> 49 watts

0.870

Table 5.8 CSL1 Energy Star 2.0 No-Load Mode Power Consumption for Product Class A1
Nameplate Output Power Maximum Power in No-Load
0 to 50watt
0.3
> 50 watts

0.5

DOE created CSL 2 to be an intermediate level between the Energy Star (CSL 1) and best
in market (CSL 3) levels (CSL3 is explained subsequently). The specific combination of no-load
power consumption and average efficiency for CSL 2 was chosen so as to optimize the tradeoff
between cost and efficiency between CSL 1 and CSL 3. To do this, DOE evaluated each
representative unit individually. DOE developed sets of efficiency and no-load power pairings
between CSL 1 and CSL 3 and estimated their resultant unit energy consumptions (UECsa) (see
chapter 7 for details on calculating UECs). DOE then compared the UECs against their
associated MSPs, which were calculated from the aggregation of manufacturer MSP data (see
chapter 5 for details on MSP aggregation). Subsequently, the final efficiency and no-load pairing
used to characterize CSL 2 was selected by examining which of the prospective pairings had the
highest weighted-average savings in the LCC analysis.
DOE then used the CSL 2 pairings for the four representative units to create equations for
average efficiency and no-load power by curve-fitting the efficiency characteristics for CSL 2
(see chapter 5 for details on curve fits). For both the average efficiency and no-load power CSL
equations, DOE used equations similar to those for CSL 1, involving linear and logarithmic
terms in the nameplate output power. DOE chose the divisions at 1 watt, 49 watts, and 50 watts
in the CSL 2 equations so that they were consistent with the nameplate output power divisions
between the equations for CSL 1. The CSL2 active-mode efficiency and no-load power
equations are shown in Table 5.9 and Table 5.10, respectively.

UEC is also referred to as Annual Energy Consumption (AEC).

5-13

Table 5.9 CSL 2 Intermediate Active-Mode Efficiency for Product Class A1


Nameplate Output Power Minimum Average Efficiency in Active Mode
(expressed as a decimal)
0 to 1 watt
0.49 * Pout + 0.15
> 1 to 49 watts
0.0701 * ln(Pout) - 0.0011 * Pout +0.647
> 49 watts
0.870
Table 5.10 CSL 2 Intermediate No-Load Mode Power Consumption for Product Class A1
Nameplate Output Power Maximum Power in No-Load
0.200 watts
0 to 50watt
0.230 watts
> 50 watts
CSL 3 reflects the most efficient products available for sale in the market (best in
market or BIM). As explained in section 5.4.3, DOE purchased EPSs using three sources in
order to identify the most efficient unit in the market. First, DOE evaluated the most efficient
units of the popular products tested in the market survey in section 5.3. Second, DOE purchased
EPS units identified in the Energy Star 2.0 databaseb. Third, DOE purchased units available
through EPS distributor websites. From among these three sources, DOE considered the best in
market EPS to be the most efficient EPSs that DOE tested, in terms of a combination of highest
average efficiency and lowest no-load power. For those units, DOE created the equation for
average efficiency using a curve-fit of the test results data, shown in Figure 5.4. DOEs
methodology for curve fitting is detailed in section 5.4.10. DOE created the equations for noload power based on the maximum no-load power among the two lower power representative
units (2.5W and 18W) and the two higher power representative units (60W and 120W),
respectively (see Section 5.4.10 for details). DOE followed this approach because there was no
clear relationship between nameplate output power and no-load power consumption. DOE chose
the divisions in the CSL 3 equations so that they were consistent with the divisions between the
equations for CSL1. Figure 5.5 illustrates the CSL3 no-load power equations, as well as the
CSL3 representative unit test and manufacturer data. The CSL3 active-mode efficiency and noload power equations are shown in Table 5.11 and Table 5.12, respectively.
Table 5.11 CSL3 Best in Market Active Mode Efficiency for Product Class A1
Nameplate Output Power Minimum Average Efficiency in Active Mode
(expressed as a decimal)
0.5 * Pout + 0.16
0 to 1 watt
0.071 * ln(Pout) - 0.0014 * Pout +0.67
> 1 to 49 watts
> 49 watts

0.8818

Taken from the Energy Star 2.0 external power supply results database in Sept 2009
http://www.energystar.gov/index.cfm?fuseaction=find_a_product.showProductGroup&pgw_code=ACD.

5-14

Table 5.12 CSL3 Best in Market No-Load Mode Power Consumption for Product Class A1
Nameplate Output Power Maximum Power in No-Load
0.10
0 to 50watt
0.21
> 50 watts
Unlike the previous CSLs, CSL4 was not based on an evaluation of the efficiency of EPS
units in the market, since EPSs with maximum technologically feasible efficiency are not
commercially available due to their high cost. Rather, to estimate the efficiency of a max-tech
unit, Navigant Consulting asked manufacturers to detail the steps they would take to achieve a
maximum efficiency for the four representative units. The interviews included manufacturers of
EPSs as well as manufacturers of integrated-circuit controllers for EPSs and manufacturer of
applications that use EPSs since those manufacturers are also intimately familiar with EPS
technologies. Navigant Consulting presented DOE with the manufacturers aggregated
responses for the representative units. The aggregated responses from manufacturers are
presented in section 5.4.5.
In contrast to other rulemakings that use a design-options approach to develop a max-tech
level, this rulemaking did not use a design option approach because of the significant variation in
the prices of individual components and their effects on efficiency. The topology and parts used
in an EPS design are typically inter-related. Hence, it is seldom possible to determine the effects
on price and efficiency of one component in isolation.
DOE verified the reasonableness of the aggregated manufacturer max-tech data before
creating curve fit equations for CSL4 To that end, DOEs subject matter experts (SMEs)
reviewed the data and confirmed it as reasonable, except for the max-tech value for the 2.5W
EPSs. The SMEs believe that 2.5W EPSs may be able to achieve a max tech efficiency of 80%
rather than the 74.0% efficiency derived from manufacturers. As noted in the executive
summary, DOE is seeking stakeholder comment on the 2.5W max tech efficiency value used.
Nonetheless, DOE created the equations for CSL4 using the aggregate manufacturer
value rather than the SMEs value because DOE had an associated cost for the aggregate
manufacturer value. As noted in the executive summary, DOE is seeking stakeholder comment
on the cost of an 80%-efficient EPS so that it can use that as the max-tech efficiency level for
2.5W EPSs in the NOPR.
Based on the representative units max-tech data, DOE created equations across all
output powers, shown in Table 5.13 and Table 5.14. DOEs methodology for curve fitting is
detailed in section 5.4.10. DOE created the equations for no-load power based on the maximum
no-load power among the two lower power representative units (2.5W and 18W) and the two
higher power representative units (60W and 120W), respectively (see Section 5.4.10 for details).
DOE chose the divisions in the CSL 4 equations so that they are consistent with the divisions
between the equations for CSL1. Figure 5.4 and Figure 5.5 illustrate the CSL4 efficiency and
no-load power equations in Table 5.13 and Table 5.14, respectively, as well as the aggregate
manufacturer data for max tech.

5-15

Table 5.13 CSL4 Max Tech Active Mode Efficiency for Product Class A1
Nameplate Output Power Minimum Average Efficiency in Active Mode
(expressed as a decimal)
0.52* Pout + 0.17
0 to 1 watt
0.079 * ln(Pout) - 0.0014 * Pout +0.67
> 1 to 49 watts
> 49 watts

0.9104

Table 5.14 CSL4 Max Tech No-Load Mode Power Consumption for Product Class A1
Nameplate Output Power Maximum Power in No-Load
0.062
0 to 50watt
> 50 watts

5.4.3

0.165

Class A EPS Testing Analysis


5.4.3.1 Survey of EPS Efficiencies in the Market

DOE purchased and tested commercially available EPS units to determine the range of
efficiencies in the market and to determine where to focus its analysis of representative units. In
the market survey (section 5.3) DOE identified 13 EPS-using applications that are important for
EPSs because they represent a large amount of shipments, consume large amounts of energy, or
both. As part of the market survey, DOE created a database of EPS models available for
purchase for each of the 13 applications. DOE purchased a number of models for each
application such that they would be roughly proportional to the shipments-weighted distributions
shown in Figure 5.2, without giving overwhelming preference to the largest shipments (e.g.,
cellular phones and notebook computers) and covering applications with the smallest shipments
(e.g., digital photo frames).
Accordingly, DOE believes that it has characterized the most popular Class A EPSs in
terms of shipments and energy consumption as shown in Figure 5.6. DOE applied the
representative unit criteria in section 5.4.1 to this survey of EPS efficiencies by application to
select representative units. Detailed explanations and results are available in Appendix 5A of
this preliminary TSD.

5-16

Market Characterization:
Output Power vs. Efficiency

EISA Level
Computer Speakers

Average EFficiency [%]

100%
90%

Cordless Phones/Answering
Devices
Digital Photo Frames

80%

External Hard Drives


Inkjet MFDs and Printers

70%
LAN Equipment
60%

Mobile Phones

50%

MP3 Speaker Docks


Notebook Computers

40%
0

50

100

VoIP Adapters

Nameplate Output Power [Watts]

Figure 5.6 Survey of EPS Efficiencies by Application


5.4.3.2 Evaluation of EPS Efficiencies at Representative Unit Values
DOE selected specific EPS units to characterize CSL0, CSL1, and CSL3 for each
representative unit. This section describes the selection process and the characteristics of the
EPSs chosen to characterize the CSLs. For each representative unit, DOE selected two to three
EPS test units on which to focus its analysis. DOE purchased test units with values that were the
same as or very close to the representative unit specifications. For the selected test units, DOE
commissioned teardowns to estimate their MPCs. For the EPSs that differed slightly from
representative unit values, DOE normalized their test results through scaling, as specified in
section 5.4.6.
The selected EPS test units were evaluated based on their scaled test results and chosen
so as to best characterize the level based on the following criteria:
All test units must adhere to the matched-pairs criteria, explained in detail
below.
The unit chosen to characterize CSL0 must meet the CSL0 requirements,
while being as close as possible to CSL0. Units meeting CSL1 criteria do not
qualify as baseline units.
The unit chosen to characterize CSL1 must meet the CSL1 requirements,
while being as close as possible to CSL1.
The most efficient unit, based on its combination of average efficiency and
no-load power consumption, characterizes and defines CSL3, Best-in-Market
(BIM).
5-17

o Units that did not meet CSL1 criteria did not qualify.
o When no single unit was dominant in both dimensions, DOE chose the
BIM unit so as not to break the matched-pairs approach.
o In addition, the CSL3 unit is chosen such that it is as far away from the
CSL1 unit as possible in the no-load power and efficiency dimensions.
o In comparing distances in the no-load power and efficiency
dimensions, the dimension with more effect on annual energy
consumption (as determined by typical application usage profiles at
that wattage level) was given precedent.
There is no test unit to characterize CSL2, Intermediate, because DOE
developed that level after finalizing test and teardown units.
There is no test unit to characterize CSL4, Max Tech, because it is a
theoretical unit that does not exist in the market.
The matched pairs approach refers to the pairings of average efficiency and no-load
power consumption that define the CSLs and EPS efficiency test results. The DOE EPS test
procedure yields two values in contrast to other DOE test procedures that only yield one value.
To develop the analysis and ultimately to set standards for EPSs DOE can either combine the test
values into a single metric, such as annual energy consumption, or it can maintain two separate
metrics. For this preliminary analysis DOE has chosen to evaluate EPSs using the two metrics
separately. DOE believes this is the most appropriate way to characterize EPSs because EPSs
have a wide variety of usage profiles which would affect any weighting of average efficiency
and no-load power. Further, this approach is important because the cost estimates from the
teardown apply to the EPS as a whole and cannot be broken down as affecting just active mode
efficiency or no-load power consumption.
DOE has structured the CSLs such that they never decrease in stringency in either metrics
and such that they always increase in stringency in at least one metric. Similarly, DOE uses
selected test units to characterize the CSLs that have matched pairs of efficiency in that as they
progress from least efficient to most efficient in terms of active mode efficiency requirements,
no-load mode power-consumption requirements, or both. As noted in the executive summary,
DOE is seeking stakeholder comment on the matched pairs approach.
DOE obtained an estimate of the manufacturer production costs for each selected test unit
used to characterize the representative units. Below, is a discussion of how DOE selected the
specific test units for each representative unit. Note that the max-tech data points in Figure 5.7,
Figure 5.8, Figure 5.9, and Figure 5.10 are only shown for context, as they are not test units,
but manufacturer responses regarding the best achievable efficiencies and no-load powers in
EPSs across representative units.
For the 2.5W, 5V representative unit, DOE considered test units within a wattage range
of 1.75W to 3W and a voltage range of 4V to 6V that met the criteria of representative product
class A1. Figure 5.7 shows a plot of the seven units that DOE considered in characterizing the
2.5W representative unit CSLs.

5-18

0.500

0.400

#876

No-Load Power [W]

Selected Test Units


0.300

Additional Test Units


Intermediate

Max Tech
CSL0
CSL1

0.200

#935

#888

CSL2
CSL3

Intermediate

#628

CSL4

#996

#994
0.100

#870

Max Tech
0.000

58%

60%

62%

64%

66%

68%

70%

72%

74%

76%

Scaled Average Active-Mode Efficiency [%]

Figure 5.7 No-Load Power vs. Scaled Efficiency for 2.5-Watt Units
DOE considered all EPS units in Figure 5.7, and ultimately selected units #876, #935,
and #996 to characterize the 2.5-watt CSL0, CSL1, and CSL3 levels, respectively. Unit #876
was chosen because it was close to the minimum CSL0 efficiency and maximum CSL0 no-load
power allowances, and was the only unit that maintained a matched-pairs approach. Unit #935
was selected because it met the CSL1 requirements, and unit #996 was the most efficient unit
found in the market that also met the CSL1 criteria. Table 5.15 shows the data for the three units
selected to characterize the 2.5W CSLs.

5-19

Table 5.15 2.5-Watt Units Used to Characterize the CSLs


Unit CSL
Nameplate
Cord
Cord
#
Output
Length Resistance
[m]
[ohms]
Power Voltage
[W]
[V]
2.4
6.0
1.86
0.31
876 CSL0

935
996

CSL1
CSL3

2.0
1.8

5.0
5.0

0.30

1.78
1.17

0.32

NoLoad
Power
[W]

Average
Scaled
Efficiency
[%]

0.400

61.4

0.170
0.103

67.9
73.5

Application

Cordless
Phones /
Answering
Devices
Generic
Mobile
Phones

For the 18W representative unit, all the units considered were exactly at the
representative nameplate output power and voltage of18W, and 12V, respectively. Figure 5.8
shows a plot of the sixteen units considered to characterize the 18W CSLs.

0.500

0.400

No-Load Power [W]

#903
Selected Test Units

#906

#949

Additional Test Units

#604

0.300

Intermediate

#118

#805
#891

Max Tech

#915

CSL0

#818

CSL1

Intermediate
0.200

CSL3

#645

CSL4

CSL2
#900

#918

#814

0.100

#809

#941

#912

Max Tech

0.000
75%

77%

79%

81%

83%

85%

87%

89%

Scaled Average Active-Mode Efficiency [%]

Figure 5.8 No-Load Power vs. Scaled Efficiency for 18-Watt Units
DOE considered all EPS units in Figure 5.8, and ultimately selected units #949, #118,
and #941 to characterize the 18-watt CSL0, CSL1, and CSL3 levels, respectively. Unit #949

5-20

was chosen because it was close to the minimum efficiency and maximum no-load power
allowances of CSL0, and was the only unit that allowed a matched pairs approach. Unit #118
was selected because it met, and was close to, the minimum efficiency and maximum no-load
power allowances of CSL1. Unit #900 was closer to the minimum CSL1 efficiency
requirements, but much farther than unit #118 in terms of no-load power from the 300mW noload power requirement. Unit #941 was chosen because it was the most energy-efficient unit on
the market. Even though unit #912 had a slightly lower no-load power, a difference on the order
of 10mW, unit #941 was over one percentage point higher in efficiency. Table 5.16 shows the
data for the three units selected to characterize the 18-watt CSLs.
Table 5.16 18-Watt Units Used to Characterize the CSLs
Unit CSL
Nameplate
Cord
Cord
#
Output
Length Resistance
[m]
[ohms]
Power Voltage
[W]
[V]

NoAverage Application
Load
Scaled
Power Efficiency
[W]
[%]

949

CSL0

18.0

12.0

1.75

0.13

0.330

78.9

Generic

118
941

CSL1
CSL3

18.0
18.0

12.0
12.0

1.90
1.81

0.11
0.11

0.270
0.100

81.4
85.4

Generic
Generic

For the 60W representative unit, the representative nameplate output power and voltage
were 60W, and 15V, respectively. DOE considered units within a wattage range of 56 to 60W,
and a voltage range of 15V to 16V. Figure 5.9 shows a plot of the seven units considered to
characterize the 60W CSLs.

5-21

0.500

No-Load Power [W]

0.400

Selected Test Units


#832

0.300

Additional Test Units


Intermediate

#855

#650

Max Tech
CSL0

#999

CSL1

0.200

#973

CSL3
CSL4

#834

CSL2

0.100

#838
Max Tech

0.000
84%

85%

86%

87%

88%

89%

90%

91%

92%

Scaled Average Active-Mode Efficiency [%]

Figure 5.9 No-Load Power vs. Scaled Efficiency for 60-Watt Units
DOE considered all EPS units in Figure 5.9, and ultimately selected units #999, #834,
and #838 to characterize the 60W CSL0, CSL1, and CSL3 levels, respectively. Unit #999 was
chosen because it was closest to the minimum CSL0 efficiency and maximum CSL0 no-load
power allowances. Unit #834 was selected because it was the only unit that met the CSL1
criteria, and simultaneously allowed unit #838, the most efficient unit on the market, to be
selected to characterize CSL3 without breaking the matched pairs approach. Table 5.17 shows
the data for the three units selected to characterize the 60W CSLs.
Table 5.17 60-Watt Units Used to Characterize the CSLs
Unit CSL
Nameplate
Cord
Cord
No#
Output
Length Resistance Load
[m]
[ohms]
Power
Power Voltage
[W]
[W]
[V]

Average Application
Scaled
Efficiency
[%]

999

CSL0

56.0

16.0

1.09

0.08

0.210

85.4

Notebook
Computers

834
838

CSL1
CSL3

60.0
60.0

15.0
15.0

1.16
1.17

0.04
0.04

0.136
0.073

87.0
88.0

Generic
Generic

5-22

For the 120W representative unit, the representative nameplate output power and voltage
were 120W, and 19V, respectively. DOE considered units within a wattage range of 119.7W to
135.1W. Figure 5.10 shows a plot of the nine units considered to characterize the 120W CSLs.

0.500

0.400

No-Load Power [W]

#947

#1002

#1006

Selected Test Units


Additional Test Units

0.300

Intermediate

#1004

Max Tech

#854

CSL0

#951

CSL1

0.200

Max Tech

CSL3

#652

#849

CSL2

#851

CSL4

0.100

0.000
84%

85%

86%

87%

88%

89%

90%

91%

92%

Scaled Average Active-Mode Efficiency [%]

Figure 5.10 No-Load Power vs. Scaled Efficiency for 120-Watt Units
DOE considered all EPS units in Figure 5.10, and ultimately selected units #854 and
#951. Unit #854 was selected to characterize the 120W CSL0 level. Unit #951was selected to
characterize both the 120W CSL1 and CSL3 levels. Unit #854 was chosen because it was
closest to the minimum CSL0 efficiency and maximum CSL0 no-load power allowances. It was
not possible to select two more units, one for CSL1, and one for CSL3 without breaking the
matched pairs approach. Therefore, unit #951 was selected to characterize both CSL1 and
CSL3, because it adhered to the matched pairs approach, and was about 1% higher in efficiency
than either unit #652 or #851. Notebook computers are the most common application that use
120-watt EPSs and they typically operate a few hours a day. For EPSs for those applications, a
1% difference in efficiency has a more significant effect on annual energy consumption than less
than a tenth of a watt in no-load power. Table 5.18shows the data for the three units selected to
characterize the 120W CSLs.

5-23

Table 5.18 120-Watt Units Used to Characterize the CSLs


Unit CSL
Nameplate
Cord
Cord
NoAverage
Application
#
Output
Length Resistance Load
Scaled
[m]
[ohms]
Power Efficiency
Power Voltage
[W]
[%]
[W]
[V]
19.0
1.78
0.05
0.230
86.5
Notebook
854 CSL0 135.1
Computers
CSL1
120.1
19.0
1.80
0.04
0.210
88.4
Notebook
951
and
Computers
CSL3

5.4.4

Class A EPS Teardowns to Estimate Manufacturer Production Cost

DOE contracted iSuppli Corp. to tear down and estimate the materials cost for select
units. DOE elected to use iSuppli for its expertise with prices in the consumer electronics
industry, since those prices are not publicly available. iSuppli provided DOE with the costs for
all parts listed in the bill of materials, and the labor for assembling those parts into an EPS. DOE
marked these costs up by the general overhead costs for running a factory to obtain the
manufacturers production cost (MPC), sometimes called the factory cost. DOE used this
information along with a markup to determine the MSPs. The following subsections describe
the process for determining MPC and the final results.
5.4.4.1 Generation of Bills of Materials
The end result of each teardown is a bill of materials (BOM). iSuppli developed BOMs
for each unit it tore down. BOMs describe each product part and the manner in which
manufacturer assembled it. The BOMs describe fabrication and assembly operations in detail,
including the process cycle times and the labor associated with each manufacturing step. The
BOM includes the following data fields for each component:
Location: The assembly/sub-assembly in which the component resides, within the device.
Quantity: The count of the component
Component Family: The general type of component such as passive and discrete
semiconductor.
Component Type: The specific type of component such as capacitor and diode.

Manufacturing Name and Part Number: The components name and number.

Component Description: Component-specific information such as Film - Radial, Dipped,

0.47uF, 10% and Zener - 34.6V, 2mA


Markings: Any visible markings used in component identification, such as Logo, 474K, n,
450MFF4 and TZX, 36, C
Package Dimensions: These include component form, diameter, length, height, width and pin
count.
Per Component Cost: The cost to the EPS manufacturer of the individual component at a
specified production volume.

5-24

Insertion Method: Either insertion by hand or automated insertion.

Per Insertion Cost (Auto): The cost of automated insertion for the component.

Hand Insert Cost: The cost of inserting the component by hand, which is calculated from the

insert time and the pay rate for the laborer.


Data Sheet Links: Any data sheets for the components used in determining pricing..
5.4.4.2 Cost Structure of the Spreadsheet Models
The manufacturing cost assessment methodology used a detailed, component- focused
technique for calculating the manufacturing cost of a product (direct materials, direct labor, and
the overhead costs associated with production). The first step in the manufacturing cost
assessment was the creation of a complete BOM from the disassembly of the units selected for
teardown. The units were dismantled, and each part was characterized according to
manufacturer, dimensions, material, and quantity. iSuppli based its assumptions on the sourcing
of parts and in-house fabrication on its industry experience and discussions with manufacturers.
The last step was to convert this information into MPC values. To perform this task, iSuppli
sums the direct material costs and the conversion costs, to which DOE added a general factory
overhead markup, to determine the total MPC for each unit. Figure 5.11 shows the general
breakdown of costs associated with manufacturing a product.
Full Cost of Product [$]

Production Cost
Direct
Labor

Direct
Materials

Overhead

Non-Production Cost
Selling

General &
Administrative

R&D

Interest

Profit

Figure 5.11 Full Cost of Product: Breakdown of Production and Non-Production Costs

5.4.4.3 Production Volumes Assumptions


A manufacturers production volumes vary depending on several factors, including
market share, the type of product produced, and if the manufacturer produces other similar
products. iSuppli estimated unit costs using three volume assumptions: annual production
volume, production lifespan, and lifetime production volume. Annual production volume
specifies the number of units a manufacturer produces of the product each year, which affects
component costs. Production lifespan specifies the number of years that a manufacturer
produces a product, which affects the amount of time over which the manufacturer can amortize
equipment costs. The lifetime production volume is the total number of units produced by the
manufacturer, which is the product of the annual production volume and production lifespan.
Based on their industry knowledge, iSuppli and DOE jointly developed estimates of
production volumes and production lifespans for each representative unit. iSuppli indicated that
their cost-estimation model was sensitive only to large differences in production volume so the

5-25

values provided were rough estimates. Furthermore, iSuppli preferred to use lifetime production
volumes for its cost estimates, whereas most manufacturers preferred annual production
volumes. As noted in section 5.4.6, DOE considered developing scaling relationships between
cost and production volume, but did not do so because it was not needed. In interviews,
manufacturers indicated to DOE that production volume, not production lifespan, was the major
factor affecting price. Consequently, the manufacturer interviews did not address production
lifespans.
5.4.4.4 Teardown Results
iSuppli performed teardowns for all of the EPSs that DOE chose to characterize the
representative units and each CSL as specified in section 5.4.3.2. For each of those EPSs, Table
5.19 indicates the estimated MPCs after applying the general factory overhead markup to
iSupplis costs.
Table 5.19 iSuppli Teardown Cost Estimates

Unit#

876
935
996
949
118
941
999
834
838
854
951
867
809
650
853
1004

Representative
Unit

CSL

MPC

2.5W
2.5W
2.5W
18W
18W
18W
60W
60W
60W
120W

Output
Cord
Cost

Output
Cord
Length
[m]

Lifespan Prod.
Volume per
iSuppli
Teardown
(Units)
6,000,000
6,000,000
6,000,000
6,600,000
6,600,000
6,600,000
7,100,000
7,100,000
7,100,000
8,000,000

CSL0
$1.42
$0.11
1.86
CSL1
$1.71
$0.17
1.78
CSL2
$1.12
$0.12
1.17
CSL0
$4.18
$0.23
1.75
CSL1
$4.11
$0.33
1.90
CSL2
$3.37
$0.25
1.81
CSL0
$4.19
$0.20
1.09
CSL1
$5.82
$0.24
1.16
CSL2
$6.15
$0.23
1.17
CSL0
$12.55
$0.65
1.78
CSL1,
120W
$9.18
$0.25
1.80
CSL2
Additional Teardowns Data (Not Used To Characterize CSLs):
*
N/A
$1.59
$0.14
1.83
18W
N/A
$3.41
$0.37
1.51
60W
N/A
$7.83
$0.44
1.75
120W
N/A
$11.53
$0.26
1.70
120W
N/A
$7.38
$0.23
1.11

8,000,000
6,000,000
6,600,000
7,100,000
8,000,000
8,000,000

*Unit #867 was a low-voltage unit, because its nameplate output voltage was 5V, and nameplate output current was
0.55A.

5.4.5

Class A EPS Manufacturer Interviews

In 2009 and 2010, on behalf of DOE, Navigant Consulting, Inc. (Navigant Consulting)
interviewed a total of eight manufacturers of EPSs, integrated circuit (IC) controllers for EPSs,
and original equipment manufacturers (OEMs) that use EPSs to obtain data on EPS efficiencies
5-26

and costs. At the request of some manufacturers, Navigant Consulting entered into non
disclosure agreements whereby it could present to DOE general information about the EPS
market and technology, but no confidential data specific to any individual manufacturer.
Navigant Consulting aggregated the manufacturer data (detailed in Appendix 5C); the resulting
cost-efficiency data is presented in section 5.4.8.
Before the interviews, Navigant Consulting gave each manufacturer an interview guide
(Appendix 5C) that included possible questions to be asked during the interview and tables
detailing efficiency and no-load values for the manufacturers to populate with associated costs.
Navigant Consulting asked manufacturers to provide feedback regarding the representation of
the market and to supply any data that could improve DOEs estimates and assumptions.
Navigant Consultings questions included the following:
1. What are the highest volume products that you sell? Please include output voltage,
output power, and application.
2. Please provide a list of any additional applications for which you sell EPSs or EPS
components.
3. Are there any specific design concerns unique to certain applications?
4. What are the typical mark ups from the EPS bill of materials (BOM) to the final

consumer purchase price?

5. The U.S. Department of Energy (DOE) is required to set no-load and active mode energy
efficiency standards for external power supplies. Is there a correlation between no-load
power and active mode power efficiency? If not, can the two be optimized separately?
6. What are your design options (e.g. Schottky diodes, improved components, better core
material) for improving active mode efficiencies and no-load power consumption?
Manufacturers provided general information and data specific to representative units. The
following subsection details Navigant Consultings methodology for aggregating manufacturer
data so that it was presentable to DOE. The subsequent subsections provide summaries of
manufacturers input on issues affecting the engineering analysis.
5.4.5.1 Aggregation Methodology
Navigant Consulting collected manufacturer cost-efficiency data for each representative
unit and aggregated manufacturer responses, which it presented to DOE. EPSs are unique
because their CSLs are defined by energy consumption parameters: average efficiency and noload power. Hence, Navigant Consulting obtained costs for each CSL defined by a combination
of EPS efficiency and no-load power. In the interview guide, Navigant Consulting asked
manufacturers for costs at the specific CSLs for each representative unit. However,
manufacturers provided costs over a range of efficiency and no-load power values because they
generally preferred to tie in costs to their own product lines, which often had efficiency and noload power values slightly different from the CSL values. Consequently, Navigant Consulting
performed three steps on the manufacturer data before providing DOE with aggregated results:
(1) normalize the data to be consistent; (2) develop equations to generalize the data; and (3)
apply the CSL values to the equations to determine aggregate costs.

5-27

Since manufacturers had used various assumptions in providing their data, Navigant
Consulting normalized the manufacturer data to ensure that the results from the manufacturers
were comparable. For example, some manufacturers had not factored in an output cord. To
account for the possible effects on cost and efficiency, the data points were adjusted, assuming
the standard 1.66m cord length. In addition, some manufacturer data assumed nameplate output
voltages or wattages that were slightly different than the representative unit values. For details
on scaling, please see Section 5.4.6.
First, the manufacturer costs which were given at the BOM or MPC points in the value
chain, needed to be marked up to MSP (per Section 5.4.6.6). Second, Navigant Consulting
ensured that the manufacturer datasets all had an MSP of $0 for baseline units. Shifting the
manufacturer data to the same CSL0 baseline values, meant that the incremental MSP costs at
CSLs above baseline, were based on the same $0 reference point. This was necessary because
scaling the datasets sometimes resulted in baseline values that were not $0.
The normalized manufacturer data was aligned at CSL0 and covered a range of values for
the other CSLs, as shown in Figure 5.12 and Figure 5.13. Both figures contain illustrative
values to show the kind of values that were obtained during manufacturer interviews. Individual
manufacturer data points cannot be revealed due to non-disclosure agreements (NDAs) with
participating manufacturers. The example is for the 2.5W representative unit. Though the
following discussion focuses on the 2.5W unit, the reasoning is the same for the 18W, 60W, and
120W representative units. CSL2 was introduced after manufacturer interviews, and hence the
costs used to characterize CSL2 were interpolated based on data from the other CSLs. Hence the
tick marks on the x-axis represent the CSL efficiency values, with the CSL2 tick mark not
corresponding to its own set of manufacturer data.

5-28

Cost vs. Efficiency


$0.35

Manufacturer Selling Price [$]

$0.30
Mfr Data Points for CSL0

$0.25
Mfr Data Points for CSL1

$0.20
Mfr Data Points for CSL3

$0.15
Mfr Data Points for CSL4

$0.10
$0.05
$0.00
55%

60%

65%

70%

75%

80%

Average Active Mode Efficiency [%]

Figure 5.12 Sample efficiency and MSP Manufacturer Data for the 2.5W Representative
Unit.
*The data presented in this plot is for illustrative purposes only. It is not actual manufacturer data.

5-29

Cost vs. No-Load Power


$0.40

Manufacturer Selling Price [$]

$0.35
$0.30
Mfr Data Points for CSL0

$0.25

Mfr Data Points for CSL1

$0.20

Mfr Data Points for CSL3

$0.15

Mfr Data Points for CSL4

$0.10
$0.05
$40

140

240

340

440

540

No-Load Power [mW]

Figure 5.13 Sample no-load power and MSP Manufacturer Data for the 2.5W
Representative Unit.
*The data presented in this plot is for illustrative purposes only. It is not actual manufacturer data.

The 2.5W CSL0, CSL1, CSL2, CSL3, and CSL4 average efficiency values were 58.3%,
67.9%, 71.0%, 73.5%, and 74.0%, respectively. The illustrative data in Figure 5.12 shows that
though the CSL1 efficiency was 67.9%, manufacturer efficiencies might vary significantly, with
efficiencies in the range of 65% to 70%. The illustrative data in Figure 5.12 does not exaggerate
the variation present in manufacturer responses. Similarly, Figure 5.13 illustrates the type of
manufacturer data that Navigant Consulting obtained for MSP versus no-load power for the
2.5W representative unit.
This set of manufacturer data presented two challenges: First, Navigant Consulting had to
account for data in three dimensions (efficiency, no-load power, and cost); Second, Navigant
Consulting had to develop an aggregate response based on the variation in the manufacturer data
along all three dimensions for each CSL of each representative unit.
To address these challenges, it was necessary to fit curves to the manufacturer data.
Curve fitting allowed Navigant Consulting to estimate an MSP at the CSL efficiency and no-load
power values, and allowed for aggregation of the manufacturer data. To explain Navigant
Consultings approach of aggregating the data in three dimensions, it is helpful to first illustrate
how the approach works in two dimensions.
Figure 5.14 illustrates the two-dimensional approach to aggregating manufacturer data.
A quadratic curve is fit to the data in each representative unit, and forced to pass through $0 at
CSL0. The linear fit is of the form Z=a+bX+cX2, where Z represents the MSP, X represents the

5-30

efficiency, and a least-squares fit determines the parameters a, b, and c. Thus, the aggregate
MSP at a particular CSL, is determined by applying the efficiency value to the equation. This is
shown in Figure 5.14 by the black circles.

Cost vs. Efficiency


$0.35

Manufacturer Selling Price [$]

$0.30

Mfr Data Points for CSL0


Mfr Data Points for CSL1

$0.25

Mfr Data Points for CSL3


Mfr Data Points for CSL4

$0.20
$0.15
$0.10

Aggregated Cost Used


to Characterize CSL

$0.05
$0.00
55%

60%

65%

70%

75%

80%

Average Active Mode Efficiency [%]

Figure 5.14 Illustration of Aggregation Method in Two Dimensions for the 2.5W
Representative Unit
*The efficiency and MSP data presented in the plot is not actual manufacturer data; it is only for illustrative
purposes.

The same approach is also applicable to no-load power and MSP values, as shown in
Figure 5.15. In this scenario, the quadratic fit would have the form Z=a+bY+cY2, where Z
would represent the MSP output, X would represent the no-load power input, and the parameters
a, b, and c would be determined from the least-squares fit.

5-31

Cost vs. No-Load Power


$0.40
Manufacturer Selling Price [$]

$0.35
$0.30
Mfr Data Points for CSL0

$0.25

Mfr Data Points for CSL1

$0.20

Mfr Data Points for CSL3

$0.15

Mfr Data Points for CSL4

Aggregated Cost Used


to characterize CSL

$0.10
$0.05
$40

140

240

340

440

540

No-Load Power [mW]

Figure 5.15 Illustration of Aggregation Method in Two Dimensions for the 2.5W
Representative Unit
*The no-load power and MSP data presented in the plot is not actual manufacturer data; it is only for illustrative
purposes.

In the original problem, the MSPs were dependent on both the efficiencies and no-load
powers. Hence, this is a three-dimensional problem where the independent variables are
efficiency and no-load power, and the dependent variable is MSP. To obtain aggregate MSPs at
the CSL efficiencies and no-load powers, Navigant Consulting fit three-dimensional quadratic
surfaces to the manufacturer data (the analog of the quadratic curves in the two-dimensional
examples). As was done in the two-dimensional examples in Figure 5.14 and Figure 5.15, one
quadratic surface was fir for each representative unit. Each quadratic surface was a least-squares
fit of the MSP, efficiency, and no-load power data, with the constraint that it pass through $0 at
CSL0. The form of the equation for the surface was Z=a+bX+cX2+dY+eY2+fXY+gX2Y+hXY2,
where Z represents the MSP output, X represents the efficiency input, Y represents the no-load
power input, and the parameters a, b, c, d, e, f, g, and h would be determined from a least-squares
fit. After fitting the surface to the manufacturer data of each representative unit, Navigant
Consulting applied the CSL efficiencies and no-load powers to obtain the aggregate MSPs.
5.4.5.2 Factors that affect conversion efficiency
Conversion efficiency is determined by EPS active-mode power consumption, which is
comprised of power lost through components as it flows through the EPS as well as power
consumed by EPS overhead control circuitry. Particularly for EPSs with higher nameplate
output powers, EPS control circuitry consumes significantly less power than what is lost as

5-32

power passes through components. Thus, manufacturers indicate that they reduce power
consumption by focusing their efforts on the power consuming components.
Design options for reducing components power consumption can be divided into two
categories: (1) improvements to the topology and IC controller and (2) improvements to
particular components in the EPS. The combination of topology and IC controller has the most
influence on EPS efficiency because the topology dictates which components are used in the EPS
and the controller coordinates how power flows. In general, manufacturers would use a flyback
topology for most EPSs, with two exceptions. For the baseline model of the 2.5W representative
unit, they would use a ringing-choke converter, which does not employ an IC controller. For the
60W and 120W representative units, they would use a resonant topology such as a half bridge
converter or an LLC (inductor-inductor-capacitor) converter for the higher-efficiency CSLs.
The IC controller dictates how power flows through the EPS by monitoring the EPS
output as well as other conditions. Using those input signals, the controller employs algorithms
to control a metal-oxide-semiconductor field-effect transistor (MOSFET) switch, which, in turn,
controls the input power that flows from mains to the primary side of the EPSs transformer. IC
controllers can use algorithms such as zero-voltage switching (ZVS) and frequency fold-back to
reduce EPS power consumption. In ZVS, the controller opens and closes the switch only when
there are 0 volts across it, which greatly reduces conduction losses in the MOSFET. A controller
employs frequency fold-back during light loading conditions by reducing the frequency at which
it turns on the MOSFET, minimizing the switching losses in the MOSFET. In addition to these
methods, the IC controller can also replace functions provided by discrete components. IC
manufacturers indicate that they typically price IC controllers based on the value of their high
efficiency and ability to replace discrete components.
Among the discrete components that manufacturers would improve to increase
efficiency, the two most often cited were the switching MOSFET and the output rectifier.
Switching MOSFETs have two types of losses: conduction losses and switching losses.
Conduction losses can be reduced by reducing the resistance from the drain to the source of the
MOSFET while turned on, referred to as the RDS_ON. Typically, increasing the size of the silicon
transformer or improving its material properties reduces RDS_ON. Switching losses are the losses
caused by activating and deactivating the MOSFET switch, typically governed by gate
capacitance. Gate capacitance is reduced through improvement of material properties and use of
smaller silicon chips. To improve MOSFETs, manufacturers typically indicated that they would
pay for larger, better quality silicon.
The output rectifier is the other key component that manufacturers would improve. There
are three types of output rectifiers: regular diodes, Schottky diodes, and synchronous rectification
MOSFETs. The power consumption through the output rectifier is governed by:
Pconsumed = Vdiode Iout. Regular diodes are the cheapest and least efficient. Typically, regular
diodes have a diode voltage drop of approximately 1V. Schottky diodes are more expensive and
have a diode voltage drop of approximately 0.3V, thus, for the same output current, their power
consumption is significantly less. A synchronous rectification circuit uses a controller and a
MOSFET instead of a diode so that the MOSFET provides the functionality of a diode without
the voltage drop. The benefit of this arrangement is that the power consumed by the MOSFET is

5-33

typically much lower than the power consumed even by a Schottky diode; however, the
MOSFET and its control circuit are more expensive than Schottky diodes. The benefit to using
synchronous rectification for EPSs becomes diminished at higher output voltages (above 15V) in
which case manufacturers sometimes found Schottky diodes to be the best design option.
As a last step in improving efficiency, manufacturers would switch to a thicker output
cable, or reduce the cable length. Changing wire gauge is generally more expensive than minor
component changes on the printed circuit board (PCB), so manufacturers first focus on the PCB
to improve efficiency at low cost. If active-mode efficiency is still below a desired level
manufacturers will change the cable to get a slight boost in efficiency, because at that point in the
design process, it is easier than redesigning the whole PCB.
5.4.5.3 Factors that affect no-load power consumption
The IC controller has the most influence over no-load power consumption because it
determines which parts of the EPS turn off in no-load mode. The most common technique
employed by IC controllers to reduce no-load power consumption is to use cycle skipping or
burst mode. In active mode, the IC controller turns on the switching MOSFET often at rates
greater than 20 kHz. In no-load mode, the controller can skip cycles, thus saving power that
would have been consumed by MOSFET switching losses. When the controller skips many
cycles possibly even for seconds at a time the EPS is said to operate in burst mode. The
number of cycles skipped is limited by the turn on time, which is how quickly the EPS needs to
be able to provide full power to the load. Better IC controllers can achieve no-load mode power
consumption in the range of 100mW to 200mW, well below the EISA maximum limit of
500mW. Typically the increase in cost is minimal to fabricate an IC controller that reduces noload power consumption; IC manufacturers charge a price for this feature that reflects its value in
the market.
For higher power EPSs (60W and 120W) manufacturers indicated another step that they
would take to reduce no-load power consumption: they would employ a small 1-watt
housekeeping power supply. In no-load mode, an EPS must monitor when there is an output
load and must meet the loads power demands in a timely manner. For higher power EPSs, the
first step to providing the monitoring function while reducing no-load power consumption is to
use an IC controller with cycle skipping and burst mode. However, that controller will still need
to activate all parts of the EPS during the bursts when it checks the presence of a load. This can
consume significant amounts of power. As an alternative, manufacturers can include an
additional housekeeping power supply circuit in the EPS that can monitor the output load and
quickly activate the main EPS when it needs to provide a load. The housekeeping power supply
will have a much smaller transformer than the one used during active mode. This reduction in
size will reduce quiescent losses. Often the housekeeping supply is a cost-effective way of
achieving very load no-load power consumption for a higher power EPS.
5.4.5.4 Factors that affect cost
Manufacturers identified the mark up chain within an EPS manufacturer as beginning as
a bill of materials (BOM) that enters a factory to a product valued at the manufacturers

5-34

production cost (MPC) as it leaves the factory to the manufacturer selling price (MSP) the
price at which the EPS manufacturer sells the EPS. Some manufacturers referred to MPC as the
factory cost since it is the cost of the EPS existing the factory, typically located in China. In
general, manufacturers provided cost data at the MPC level and indicated a typical mark up of
about 1.3 from MPC to MSP, although the range of markup varied from 1.2 to 1.85.
Manufacturers also indicated that markups varied with efficiency i.e., EPSs qualifying for
Energy Star had higher markups. Nonetheless, the analysis only uses the markups for baseline
EPSs because that best reflects what consumers would pay were DOE to implement a standard.
In addition, manufacturers provided data on cost scaling. They unanimously indicated
that cost scaling is unnecessary between EPSs with similar nameplate output power values
(within approximately 10% of each other). Specifically, many manufacturers provided data for
65W EPSs, which they believe are more common than 60W EPSs, the representative unit.
Nonetheless, they believe that only a few of the components between the two EPSs would be
different and the cost difference between those components is negligible.
In other cases, manufacturers provided data for units that were significantly different from the
representative unit values. One manufacturer provided data for a 5W EPS and indicated that the
costs for a comparable 2.5W EPS would be half. Another manufacturer provided data for a
100W EPS and indicated that the costs for a 120W EPS would be 10% higher.
5.4.5.5 Factors to consider for candidate standard levels
Manufacturers had a consensus view that using matched pairs as the basis for the CSLs is
a valid approach. Specifically, they indicated that average efficiency and no-load power
consumption vary independently, except at very high efficiencies i.e., when average efficiency
is very high (above 90% for EPSs over 50W) or no-load power consumption is very low (less
than 30mW).
5.4.5.6 Factors to consider when interpreting the test and teardown results
Manufacturers have a consensus view that cost increases with efficiency when all other
characteristics of EPS design are held constant; hence, the manufacturers cost-efficiency curves
all have positive slopes. They noted, however, that conducting a cost-efficiency analysis by
purchasing EPSs might yield questionable results because it is all but impossible to hold constant
all design characteristics, other than cost and efficiency, for different EPSs in the market.
Example EPS characteristics that they noted might vary:
Maximum case temperature
Maximum component temperatures
Component de-rating (how close a component operates to its maximum rating)
Hold-up time (how long the EPS outputs power after being disconnected from mains)
Output voltage regulation
Efficiency requirements at non-US voltages
Protection features (e.g., shutting down during short circuit conditions)
Maximum ambient temperature

5-35

Most of these characteristics affect the reliability of the EPS (i.e., when it fails) and the
tolerance of the EPS to different electronic and environmental conditions (e.g., how long the EPS
provides output power after being unplugged from mains; whether the EPS operates in 110
degree heat). A further complication is that many of these factors significantly affect cost and
efficiency, but do not all affect it in the same way i.e., characteristics leading to more reliable
and expensive EPSs may make EPSs more efficient in some cases, but less efficient in others.
For instance, increasing component de-rating makes an EPS more expensive and reliable, but
less efficient. De-rating is the value for a component, such as a transistor, that specifies the
difference between the anticipated maximum voltage the transistor will experience under normal
operating conditions in the EPS versus the maximum rated value indicated by the transistor
manufacturer. Often transistors in EPSs are expected to endure 400V drops and will often have a
maximum rating of 500V or 600V. Of course, the 600V transistor is more expensive, but it is
also less efficient than the 500V one. So if two EPSs were identical other than their transistor
de-ratings, their cost-efficiency curve would have a negative slope. However, an EPS with a
600V transistor will be more robust. In other cases, a higher-quality specification, such as
lowering case temperature, would tend to lead to improved efficiency. Ultimately, the OEM
specifies these characteristics so that the EPS functions as desired with the OEMs application or
range of applications. EPSs are used with a diversity of applications; thus, their specifications
are diverse, which complicates an analysis of cost and efficiency based on commercially
available units.
To deal with this issue in the test and teardowns section of the preliminary analysis, DOE
notes the intended application for the EPS test units selected to characterize the representative
units. DOE also had its SME conduct evaluations to compare the selected test units. In
particular, the SME focused on the more important characteristics that manufacturers mentioned.
For future analyses, manufacturers suggested purchasing EPSs from similar manufacturers for
similar types of products. For example, for the 60-watt representative unit analysis, DOE might
only buy EPSs sold with laptops from major manufacturers since they tend to have similar
requirements. Currently, the 60-watt representative unit analysis includes test units for generic
applications that might be lower quality than laptop EPSs. However, even if DOE were to
narrow the unit selection, manufacturers still expect that differences in the EPS specifications
between similar OEMs could make it difficult to isolate cost and efficiency so as to yield a costefficiency curve with a positive slope. In addition, DOEs SMEs indicate that the redesign
process of an EPS represents a fixed cost which the test-unit teardown analysis cannot account
for.
5.4.6

Class A EPS Scaling Relationships

DOE developed scaling relationships that it used both to analyze representative unit data
as well as to scale CSLs from product class A1 to other product classes. In general, the scaling
methods for representative unit data were detailed whereas the scaling methods for CSLs were
simpler. For the representative unit data DOE adopted the more detailed approach because the
detailed methods provide enhanced accuracy, which was paramount, as it was the basis for later
analyses, including the LCC and NIA. For the CSL equations, DOE adopted the simpler
approach so that stakeholders could more easily evaluate the CSLs. As noted in the executive

5-36

summary, DOE is seeking stakeholder comment on these approaches and may develop more
detailed CSL equations in response to stakeholder comment.
For the purposes of the preliminary analysis, DOE developed CSL equations that have
the general form Y=a*ln(Pout) + b*Pout+c, for each of the nameplate output power segments,
where Y indicates the efficiency or no load-power requirement; Pout indicates the nameplate
output power; and a, b, and c indicate the specific parameters defined in the respective CSLs. As
explained in this section, DOE has determined that output cord and output voltage also have an
effect on efficiency. Thus DOE could have included terms in the CSL equations for output cord
and output voltage although that would have resulted in much more complicated equations. For
the NOPR, DOE intends to maintain the equations in this form, which is based on output power
exclusively.
In sections 5.4.3, 5.4.4, and 5.4.5 DOE presents the test data, teardown data, and
manufacturer data that it used as the basis of this analysis. Wherever possible, DOE selected test
units that had the same characteristics as the representative units. Similarly, it sought data from
manufacturers at the representative unit values. However, in some cases, test units and
manufacturers data were slightly different from the representative unit specifications. For those
cases DOE developed detailed scaling relationships so that the data would be most applicable to
the representative unit analysis.
Specifically, the Class A EPS engineering analysis characterizes the cost-efficiency
relationship using average efficiency in active mode, power consumption in no-load mode and
MSP. DOE did not develop scaling relationships for no-load mode power consumption because
it is minimally affected by EPS representative unit characteristics. DOE considered developing a
scaling relationship for cost by production volume. Specifically, for the representative units,
DOE and iSuppli, the contractor that DOE employed to carry out the teardowns, developed
estimate production volumes that were on the order of approximately 1M units shipped per year.
iSuppli indicated that prices did not change appreciably over that range, hence scaling by
production volume is not necessary. Thus, to scale data with different production volumes to the
representative units production volume, DOE did not change the datas costs, which DOE
considers to be scaling with no effect.
For the efficiency and MSP characteristics, DOE developed the following scaling
relationships:
Efficiency by nameplate output power. DOE scaled efficiency by output power
using the EISA 2007 standard and Energy Star 2.0 specification, which follow the
market trend of increasing efficiency with increasing output power. (Section
5.4.6.1).
Efficiency by nameplate output voltage. For EPSs of a given nameplate output
power, lower nameplate output voltage results in higher current-associated losses
and lower efficiency. DOE analyzed units with the same nameplate output power
and different nameplate output voltages to characterize this relationship. DOE
also evaluated a voltage scaling approach based on Energy Star 2.0, for lowvoltage units (Section 5.4.6.2).

5-37

Efficiency by output cord length. Longer output cords provide consumers with
the added utility of being able to operate the product farther from mains. Longer
output cords also have higher losses and are more expensive. DOE analyzed
output cords to characterize this relationship. (Section5.4.6.3).
Cost by nameplate output power. EPSs with higher nameplate output powers
provide consumers with the added utility of being able to provide power to more
power-demanding products. EPSs with higher nameplate output power also have
more and larger components and, consequently, cost more. (Section 5.4.6.4).
Cost by output cord length. Longer output cords are also more expensive. DOE
analyzed output cords to characterize this relationship. (Section 5.4.6.5).
Markups. Test and manufacturer data costs at different points in the value chain
represent different stages in the sale of an EPS. DOE characterized these markups
in order to compare costs at the same stage in the value chain, namely MSP
(Section 5.4.6.6).
Though in many cases the data did not require applying all of the scaling steps, all the
scaling steps are nonetheless illustrated in Figure 5.16 in the order that they would be applied.

Original
Efficiency

Original Cost

Scaling for
Cord Length

Scaling for
Cord Length

Scaling for
Output
Power

Scaling for
Output
Voltage

Scaling for
Output
Power

Markups

Scaled Efficiency
and Cost

Figure 5.16 Scaling Steps to Normalize Efficiencies and Costs


5.4.6.1 Scaling Efficiency with Output Power
The practically achievable efficiency of an EPS depends on its nameplate output power,
with lower-power EPSs tending to exhibit lower active-mode efficiencies than their higherpower counterparts. DOE characterized this relationship using the EISA 2007 standards
equation and ENERGY STAR 2.0 standard voltage specification equation that describe this
market trend; these equations are the same as CSL0 and CSL1 equations, shown in Table 5.5
and Table 5.7, respectively.
DOE used these equations as references relative to which it scaled data. To scale the
data, DOE ensured that the ratio between the original data point and the reference equations
remained constant, as shown in Figure 5.17. The figure shows how the efficiency data point
maintains the same relative relationship between the EISA 2007 and ENERGY STAR 2.0
equation as it is scaled from 70% efficient at 5 watts to 75% efficient at 10 watts. DOE believes
this scaling approach is appropriate because it ensures that the EPS data does not cross CSLs as

5-38

it scales between output powers. Further, by comparing the data relative to two levels, rather
than shifting based on the absolute difference from one level, there is a lower risk of the data
being scaled to unrealistic values.
EISA 2007

ENERGY STAR 2.0

85%

Shift Output Power

Efficiency [%]

80%

New
value

75%
Original
value

70%
65%
60%
55%
0

10

12

14

Nameplate Output Power [W]

Figure 5.17 Scaling an EPS efficiency data point with nameplate output power
DOE developed the following equation to perform the scaling:
UNIT _ ORIGINAL
UNIT _ NEW

EISA _ ORIGINAL

EISA _ NEW

E*_ NEW
E *_ ORIGINAL

EISA _ NEW

EISA _ ORIGINAL

Where UNIT , EISA , and E* are the efficiencies of the unit under analysis at an output power,
the corresponding EISA 2007 efficiency, and ENERGY STAR efficiency, respectively.
ORIGINAL and NEW denote whether the values are at the units original nameplate output
power or new nameplate output power.
In certain instances scaling by output power does not have an effect. For data where both
the ORIGINAL nameplate output power and the NEW output power are greater than or equal to
51W, there will be no effect on the efficiency value. This is because both the EISA and Energy
Star 2.0 equations are constant for output powers at or above 51W. Hence, the efficiency value
output from the scaling equation will be identical to the efficiency value used as the input.
Unit #876 is an example of an instance where it was necessary to scale based on output
power. Unit #876 was chosen for the 2.5W CSL0 representative unit (see Table 5.15). Since
this unit had a nameplate output power of 2.4W, it was necessary to scale its efficiency to that of

5-39

an equivalent 2.5W unit. The original, tested average active-mode efficiency of unit #876, which
was 61.0%, was first scaled using the output cord scaling procedure (see Section 5.4.6.3), to
61.1%. Then, after applying the output power scaling, the efficiency was scaled from 61.1% to
61.4%. The magnitude depends in general on how close the original nameplate output power is
to the representative output power and the slope of the curves at those values.
5.4.6.2 Scaling Efficiency with Output Voltage
DOE used two methods for output voltage scaling. The first method was simple and
based Energy Star 2.0. The second method was more detailed and based on test data. DOE
applied the first method to scale CSL equations between the basic voltage and low-voltage
product classes. DOE applied the second method to scale data for the representative unit
analysis. As noted in the executive summary, DOE is seeking stakeholder comment on this
approach. Herein DOE describes the two methods for output voltage scaling in more detail.
The first method involved using the differences between the EPS Energy Star 2.0
efficiency equations for basic and low-voltage (see Section 5.4.10.2 for details). This method
was used in scaling CSL efficiency equations from the basic voltage product classes (A1 and A3)
to the low-voltage product classes (A2 and A4). This method was particularly appropriate for
product class scaling for at least two reasons. First, the low-voltage product class definitions
coincide with the Energy Star 2.0 definitions for basic voltage and low voltage. Second, using
the Energy Star 2.0 equations, which span the entire range of nameplate output powers, allowed
DOE to set a standard for efficiency at all nameplate output powers, not just those near the
representative unit values, where the test data method is focused.
DOE believes that the Energy Star 2.0 scaling method was inappropriate for use directly
on representative unit data. Specifically, unlike the Energy Star 2.0 standard voltage equation,
the Energy Star 2.0 low-voltage efficiency equation applies to EPSs with output voltage less than
6V, and output current greater than or equal to 0.55A. Thus, as the Energy Star 2.0 standard
voltage and low-voltage labels imply, the Energy Star 2.0 efficiency equations only provide a
coarse, two-bin, basis for examining the effects of differing output voltages on EPS efficiencies.
Thus, this first method was not applicable for scaling the representative unit data to the
appropriate representative unit voltages at fine scales.
DOE developed the second, more complex method based on units it had tested in the
market near representative unit values. This second method was particularly useful for scaling
representative unit data by output voltage. Although DOE could theoretically extend this method
to develop scaling relationships at other nameplate output powers, DOE did not do so because of
the significant complexity involved. In addition, even if DOE were to extend this method to all
output powers, it would still not address the issue of how to scale a basic voltage product class to
a low voltage product class. This is because DOEs output voltage scaling method requires an
exact target output voltage to scale to, and the low-voltage product class only indicates that the
output voltage is less than 6 volts. The following is an introduction to this second method, which
DOE used to scale representative unit data by output voltage:

5-40

EPS power consumption is related to (1) power consumption due to overhead circuitry
and (2) losses as power flows through the EPS. For an EPS of a given output power, overhead
circuitry consumes power independent of EPS output voltage. In contrast, losses as power flows
through the EPS are directly related to output voltage, because the combination of output voltage
and output power determines output current. For instance, resistive losses are related to output
current by I2*R; losses due to diode drops are related by I*Vdiode.
DOE analyzed this relationship for each of the four representative units by analyzing EPS
product families. Manufacturers create EPS product families by designing an EPS for a specific
nameplate output power and a range of nameplate output voltages. Consequently, product
families are comprised of EPSs that are very similar, only differing in output voltage. DOE
purchased and tested (using the DOE EPS test procedure) all of the EPS product families it could
locate with nameplate output power close to or at the representative unit output powers. For
those product families with different nameplate output powers, DOE scaled the efficiency
results, per section 5.4.6.1.
By analyzing the EPS product families DOE determined a scaling relationship between
output voltage and efficiency for all of the representative units, except the 2.5-watt representative
unit. DOE believes that 2.5-watt EPSs do not have a scaling relationship because these EPSs
have very low nameplate output power such that the overhead losses are a very significant
portion of the total losses. Thus the overhead losses obscure the changes in power consumption
losses that vary with output voltage, so no scaling with output voltage is necessary.
For the remaining three representative units, DOE developed a low-efficiency curve and
a high-efficiency curve to characterize the lower-efficiency and higher-efficiency product
families. The following discussion illustrates how DOE developed a low-efficiency and highefficiency curve for the 60-watt representative unit. The same methodology applies to both the
18-watt and 120-watt representative units.
To perform output voltage scaling for 60-watt EPSs, DOE analyzed six manufacturer
product families, as shown in Figure 5.18.

5-41

92%

Mfr F

90%

Efficiency [%]

Mfr D
88%
Mfr B
Mfr G

86%

Mfr C
84%
Mfr H

82%

High
Low

80%
0

10

20

30

40

50

60

Output Voltage [V]

Figure 5.18 60-Watt EPS Product Families for Output Voltage Scaling
DOE identified Manufacturer Bs product family as having the highest efficiencies, and
thus used its data points to create a best fit logarithmic curve to represent the higher bound for
efficiency. Figure 5.19 presents Manufacturer Bs family product data points and the higher
bound best fit logarithmic curve.

92%

90%

Efficiency [%]

Mfr B
88%

86%

High

84%

82%

Log. (Mfr
B)

80%
0

10

20

30

40

50

Output Voltage [V]

Figure 5.19 High-Efficiency Curve Used for Output Voltage Scaling

5-42

60

To create the lower bound for efficiency, DOE considered using Manufacturer Ds
product family. However, DOE rejected Manufacturer Ds product family because it would
cause the low-efficiency curve to be too close to the high-efficiency curve, which would limit the
effectiveness of the scaling relationship. Instead, DOE used a combination of Manufacturer F
and Manufacturer Hs product families as shown in Figure 5.20. DOE rejected Manufacturer Fs
first two points because they do not form a consistent trend with the rest of the family. DOE
rejected Manufacturer Hs first point because its efficiency (84.1%) is below the EISA level.
DOE did not include in its analysis any data points that did not meet EISA standards.
89%
88%

H4

Efficiency [%]

87%

F1

F4

H3

F2

Mfr F

F3

86%
H2

85%

Mfr H

H1

84%
83%
82%

Low

81%

80%
0

10

20

30

40

50

60

Output Voltage [V]

Figure 5.20 Product Families Used to Create Low-Efficiency Curve for Output Voltage
Scaling
DOE created a best-fit logarithmic curve to represent the lower bound for efficiency
using Manufacturer Hs second data point and Manufacturer Fs third and fourth data point,
shown in Figure 5.21.

5-43

89%
Mfr F
88%
Mfr H
Low

Efficiency [%]

87%

86%

85%

84%

83%

82%
0

10

20

30

40

50

60

Output Voltage [V]

Figure 5.21 Low-Efficiency Curve Used for Output Voltage Scaling


In summary, Figure 5.22 presents the data points used to create the best fit curves, as
well the upper and lower bounds for efficiency.
0.92

y = 0.0308ln(x) + 0.7918
R = 0.8746

0.91

High
Low

0.9

Log.
(High)

Efficiency [%]

0.89
0.88
0.87
y = 0.0203ln(x) + 0.7979
R = 0.895

0.86
0.85
0.84
0.83

0.82
0

10

20

30

40

50

60

Output Voltage [V]

Figure 5.22 High-Efficiency and Low-Efficiency Curves for Output Voltage Scaling
Table 5.20 presents the equations for the low-efficiency curves and a high-efficiency
curves relating efficiency (Eff) to nameplate output voltage (Vout) for each representative unit.
The results for the four representative units are shown in Figure 5.23, Figure 5.24, Figure 5.25,
and Figure 5.26. Where voltage scaling was necessary, DOE scaled test unit and manufacturer
5-44

data relative using these equations for output voltage in the same manner it used the EISA 2007
and ENERGY STAR 2.0 equations for output power, discussed in section 5.4.6.1. As noted in
the executive summary, DOE is seeking stakeholder comment on its voltage scaling procedure.
Table 5.20 Curves Characterizing the Relationship between Output Voltage and Efficiency
Representative
Low-Efficiency Curve
High-Efficiency Curve
Unit
18 W
Eff = 0.0371 * ln(Vout) + 0.6997
Eff = 0.0143 * ln(Vout) + 0.8092
60 W
Eff = 0.0203 * ln(Vout) + 0.7979
Eff = 0.0308 * ln(Vout) + 0.7918
120 W
Eff = 0.0203 * ln(Vout) + 0.8086
Eff = 0.0248 * ln(Vout) + 0.8061
74%
72%

Efficiency [%]

70%
Mfr B (scaled 3W)

68%
66%

Mfr B (scaled 2W)

64%
62%
60%
0

10

12

Output Voltage [V]

Figure 5.23 2.5-Watt EPS Product Families for Output Voltage Scaling

5-45

14

88%
Mfr A

86%
Mfr B

Efficiency [%]

84%

Mfr C

Mfr D

82%

Mfr E

80%
High

78%

Low

76%

Log.
(High)
Log.
(Low)

74%
0

10

20

30

40

50

60

Output Voltage [V]

Figure 5.24 18-Watt EPS Product Families for Output Voltage Scaling

92%
Mfr F

90%

Mfr D

Efficiency [%]

Mfr B
88%
Mfr G

Mfr C

86%

Mfr H
High

84%

Low
82%

Log.
(High)
Log.
(Low)

80%

10

20

30

40

50

Output Voltage [V]

Figure 5.25 60-Watt EPS Product Families for Output Voltage Scaling

5-46

60

92%
Mfr B

90%

Efficiency [%]

Mfr A
88%

Mfr D
Mfr I

86%

Mfr H
84%

Mfr J
High

82%

Low
80%

Log.
(High)
Log.
(Low)

78%
0

10

20

30

40

50

60

Output Voltage [V]

Figure 5.26 120-Watt EPS Product Families for Output Voltage Scaling
An example of an instance where output voltage scaling was necessary was with unit
#999 which had a nameplate output power of 56W and voltage of 16V. Unit #999 was used as
the 60W CSL0 representative unit (see Table 5.17). Since this unit had a nameplate output
voltage of 16V, it was necessary to scale it to that of an equivalent 15V unit. The original, tested
efficiency of unit #999 was 86.0%, which was first scaled using output cord (see Section 5.4.6.3
for details) and output power scaling (output power scaling has no effect between 56W and
60W see Section 5.4.6.1 for details) to 85.5%. Then, after applying the output voltage
scaling, the efficiency was scaled from 85.5% to 85.4%.
5.4.6.3 Scaling Efficiency with Output Cord
The output cord of an EPS can have an appreciable impact on its measured efficiency due
to resistive losses in the conductors. Based on test unit data, the output cord, which was
determined to be of length 1.66m, can cause an efficiency drop of up to 1.97 percentage points
versus units with no cord. Table 5.21 below shows the average and maximum efficiency drops
between a unit with no output cord and a unit with a 1.66m cord, for each of the representative
units:

5-47

Table 5.21 Effects of a 1.66m Output Cord on EPS Efficiency in Percentage Points
Average Efficiency Maximum Efficiency
Drop
Drop
Representative Unit
[Percentage Points]

[Percentage Points]

2.5W

0.59%

0.73%

18W

0.63%

0.98%

60W

0.82%

1.16%

120W

0.84%

1.97%

DOE considered the effect of the output cord significant in creating CSLs for the Class A
EPSs engineering analysis, based on the fact that merely a few percentage points separate the
most and least efficient units on the market at a given output power level. Additionally,
differences in cord length impact consumer utility: a longer output cord, which is less efficient,
provides additional consumer utility by increasing the reach between the EPS and the end-use
application. Figure 5.27 shows a plot of the kinds of output cord lengths found in the units DOE
tested.

5-48

Cord Lengths of Tested EPS Units

16

Number of Units

14
12
10
8
6
4
2

2.03

1.98

1.92

1.90

1.88

1.86

1.84

1.82

1.80

1.78

1.76

1.73

1.69

1.63

1.52

1.50

1.45

1.22

1.18

1.16

1.10

0.95

0.91

Cord Length (m)


Figure 5.27 Cord Lengths of Tested EPS Units
*Data as of March 5th, 2010.

Since there are so many different cord lengths, DOE normalized tested efficiencies based
on a single baseline cord length for the preliminary analysis. The 1.66m average cord length,
was derived from the individual average cord lengths in each representative unit, as shown in the
table below:
Table 5.22 Average Output Cord Lengths
Representative
Unit
2.5W
18W
Average Cord
Length [m]

1.77

1.78

60W

120W

1.52

1.55

Average of
Averages
1.66

*Data as of March 5th, 2010.

To derive the output cord efficiency scaling equation, DOE used the fact that the
electrical resistance R of a wire depends on the resistivity
, length
, and cross sectional
area
, of the wire, in the following manner:
. Therefore, an identical EPS whose

5-49

output cord length was hypothetically changed from length , to a baseline-length


LB
an output cord resistance of RNORM
RCORD .
L

would have

Secondly, the power into an EPS is partially transmitted out of the output cord end and
partially consumed in the printed circuit board and the output cord components. The third fact
that was used is that the power loss in the output cord is equal to the square of the current times
the resistance of the cord. The second and third equations are shown below:
PIN

PPCB

PCORD

POUT

I 2 RCORD

PCORD

Using these three pieces of information, the output cord efficiency scaling equation was
derived. The equation for modifying the efficiency of an EPS with a certain output cord length
to that with a different output cord length is shown below:
POUT
NORM

L
I B RCORD
L
2

PIN

POUT
2

I RCORD

PIN

I RCORD

LB
L

.
1

Variables Defined
PIN : Input power

PPCB

: Power consumed by the EPS device in the conversion process

PCORD

: Power loss due to resistance in the cord

POUT

: Output power

: Output current

RCORD

: Resistance of a given cord

R NORM
CORD

NORM

: Resistance of a cord normalized for length

: Efficiency of a given cord


: Efficiency of a given cord, normalized for length

L B : Baseline cord length, defined to be 1.66m in this analysis.

5-50

As noted in the executive summary, DOE is seeking stakeholder comment on its output
cord scaling procedure.
One example of an instance where it was necessary to scale based on the length of the
output cord was with unit # 999, which was chosen for the 60W CSL0 representative unit (see
Table 5.15). Since this unit had an output cord of length 1.09m, it was necessary to scale its
efficiency to that of a unit with a 1.66m cord. The original, tested average active-mode
efficiency of unit #999, which was 86.0%, was therefore scaled to 85.5% using the output cord
scaling procedure. In general, the effect could be smaller or larger depending on how far away a
units output cord is from 1.66m, and the resistance of the output cord per unit length.
Output cord scaling was used to normalize the representative unit data points; however the CSL
equations do not consider output cord length. As stated previously, DOE did not include output
cord length in the CSL equations to maintain their simplicity. Instead, the CSL equations are
developed based on the assumption that all EPS output cords are 1.66m. As noted in the
executive summary, DOE is seeking stakeholder comment on whether to include output cord
length in the CSL equations.
5.4.6.4 Scaling Cost by Nameplate Output Power
In interviews, manufacturers indicated that for products whose nameplate output power is
close to the representative unit output power (within 10%), the cost difference is negligible,
hence there is no need to scale cost by nameplate output power. However there were a few
exceptions: one manufacturer provided data for a 5-watt EPS to characterize the 2.5W
representative unit. In that case, the manufacturer indicated that costs should be divided in half,
consistent with the difference in output power. Similarly, another manufacturer characterized the
120-watt representative unit using a 100-watt EPS; the manufacturer indicated that the costs
should be scaled proportionally.
5.4.6.5 Scaling Cost by Output Cord Length
As noted previously, DOE scaled efficiency by output cord length for test unit.
Similarly, DOE scaled MSP by output cord length for selected test units that iSuppli tore down
whose cord lengths were different from the representative units cord length of 1.66m. For the
test units, DOE scaled the cost of the output cord in the BOM proportionally to the change in
length.
For the manufacturer data, some manufacturers noted that they did not account for cord
losses in their cost-efficiency data. For those units, DOE lowered the efficiency of the units, but
did not change their MSPs since the all MSPs were provided on a relative scale. Consequently,
increasing all MSPs by the same amount did not affect the relative relationship between the
costs.
5.4.6.6 Class A EPS Markups

5-51

DOE gathered inputs on markups from manufacturer interviews. Specifically, DOE


questioned manufacturers regarding typical markups for an EPS between MPC, the cost of the
EPS as it leaves the factory, and MSP, the price at which the EPS is sold to an OEM. DOE
aggregated the data provided by multiple manufacturers to determine that a typical markup from
MPC to MSP is approximately 35.5%, and a typical markup from BOM to MSP is 62.5%. DOE
used this markup to determine MSP for selected test units as well as manufacturer data. Figure
5.28 illustrates the markup chain from the original parts to the end-consumer.
Components

EPS Manufacturer
Original Device Manufacturer (ODM)

Original Equipment
Manufacturer
(OEM)

Consumer

Integrated
Circuit (IC)

Others
Labor and Utilities

Bill of
Materials
(BOM)

Manufacturer
Production Cost
(MPC or Factory
Cost)

Manufacturer
Selling Price
(MSP)

Retail Price

Figure 5.28 The Full Markup Chain, including the steps from BOM to MPC to MSP.
5.4.7

Class A EPS Cost and Efficiency Relationships from Testing and Teardowns

For each representative unit, DOE combined the scaled efficiency test results with the
scaled MSP for the three EPS test units used to characterize the baseline efficiency level, CSL1,
and CSL3. As explained in section 5.4.3.2, DOE selected test units to characterize CSL0, CSL1,
and CSL3 such that the units scaled efficiencies (per section 5.4.6) were equal to or slightly
more efficient than the CSLs detailed in section 5.4.2 and that the units efficiencies maintained
the matched-pairs relationship. For each of these test units, iSuppli conducted teardowns to
estimate MPC, as explained in section 5.4.4. DOE then scaled and marked up the test units
MPC values to MSP values.
DOE used the first eleven test units presented in Table 5.19 to characterize the CSLs for
the four representative units. Before DOE used the data to generate cost-efficiency relationships,
DOE normalized the data through efficiency and cost scaling steps. The scaling steps were
applied in the order shown in Figure 5.16. Table 5.23 shows the details of which test units
underwent which types of scaling. For details on the scaling procedures, see Section 5.4.6. For
some of the test units that underwent scaling, some of the scaling had no effect. Specifically,
output voltage scaling had no effect on 2.5W representative units (see Section 5.4.6.2 for
explanation), and output power scaling had no effect on the 60W and 120W representative units
(see Section 5.4.6.1 for explanation). Following Table 5.19, DOE presents the test and teardown
5-52

results in Table 5.24, Table 5.25, Table 5.26, and Table 5.27 for the 2.5W, 18W, 60W, and
120W representative units, respectively.
Table 5.23 Scaling and Markup Steps Performed On The Test Units
Efficienc Efficienc
Cost
Rep.
Efficienc
Cost
Test y Scaling y Scaling
Scaling
Unit
y Scaling
Scaling
Mark-Up
Unit
by
by
by
Output
by Cord
by Cord
to MSP
#
Output
Output
Output
Power
Length
Length
Power
Voltage
Power

876
*

935
2.5W

996

949

18W
118

941

999
*

834
60W
838

854
*
120W

951
*
Indicates that the data was scaled to the representative unit values, with effects on
efficiency or cost.

* Indicates that the data was scaled to the representative unit values, although the scaling had

no effect.

Table 5.24 Cost and Efficiency Relationship for 2.5W EPS (Testing and Teardowns)
CSL 0
CSL 1
CSL 3
CSL 4
Test Unit Efficiency [%]:
61.4%
67.9%
73.5%
N/A
Test Unit No Load Power [W]:
0.400
0.170
0.100
N/A
Energy Star
Best in
CSL Description:
EISA
Max Tech
2.0
Market
(58.3%,
(73.2%,
(73.9%,
CSL Eff. [%], No-Load Power [W]
(67.9%, 0.300)
0.500)
0.100)
0.062)
MSP Difference with CSL0 [$]:
$0.00
$0.39
-$0.32
N/A
Cordless
Phones
Generic
Mobile
Phones
Original Application:
N/A
Test Unit #:
876
935
996
N/A

5-53

Table 5.25 Cost and Efficiency Relationship for 18W EPS (Testing and Teardowns)
CSL 0
CSL 1
CSL 3
CSL 4
Test Unit Efficiency [%]:
78.9%
81.4%
85.4%
N/A
Test Unit No Load Power [W]:
0.330
0.270
0.100
N/A
CSL Description:
EISA
Energy Star 2.0 Best in Market Max Tech
(87.3%,
CSL Eff. [%],No-Load Power [W] (76.0%, 0.500) (80.3%, 0.300) (85.0%, 0.100)
0.062)
MSP Difference with CSL0 [$]:
$0.00
-$0.12
-$1.10
N/A
Generic
Generic
Generic
Original Application:
N/A
Test Unit #:
949
118
941
N/A

Table 5.26 Cost and Efficiency Relationship for 60W EPS (Testing and Teardowns)
CSL 0
CSL 1
CSL 3
CSL 4
Test Unit Efficiency [%]:
85.4%
87.0%
88.0%
N/A
Test Unit No Load Power [W]:
0.210
0.136
0.073
N/A
CSL Description:
EISA
Energy Star 2.0 Best in
Max Tech
Market
CSL Eff. [%], No-Load Power
(85.0%,
(87.0%, 0.500) (88.0%,
(91.0%,
[W]
0.500)
0.210)
0.165)
MSP Difference with CSL0 [$]:
$0.00
$2.20
$2.64
N/A
Notebook
Generic
Generic
Original Application:
N/A
Computers

Test Unit #:

999

834

838

N/A

Table 5.27 Cost and Efficiency Relationship for 120W EPS (Testing and Teardowns)
CSL 0
CSL 1
CSL 3
CSL 4
Test Unit Efficiency [%]:
86.5%
88.4%
88.4%
N/A
Test Unit No Load Power [W]:
0.230
0.210
0.210
N/A
CSL Description:
EISA
Energy Star
Best in
Max Tech
2.0
Market
CSL Eff. [%], No-Load Power
(85.0%,
(87.0%, 0.500) (88.0%,
(91.0%,
[W]
0.500)
0.210)
0.165)
MSP Difference with CSL0 [$]:
$0.00
-$4.53
-$4.53
N/A
Notebook
Notebook
Notebook
Original Application:
N/A
Test Unit #:
5.4.8

Computers

Computers

Computers

854

951

951

N/A

Class A EPS Cost and Efficiency Relationships from Manufacturer Interviews

For each representative unit, DOE combined the scaled efficiency test results with the
MSP for the three EPS test units used to characterize the baseline efficiency level, CSL1, and
CSL3. The CSL values for CSL4 came from manufacturer interviews. Table 5.29, Table 5.30,
Table 5.31, and Table 5.32 list data for each CSL for each representative unit.

5-54

Similar to the test units, the manufacturer data underwent a number of the efficiency and cost
scaling steps. The scaling steps were applied in the order shown in Figure 5.16. Table 5.28
shows the details of which manufacturer representative units underwent which types of scaling.
Though most of the individual manufacturer data was not scaled, Table 5.28 indicates scaling
wherever data for at least one manufacturer was scaled. Individual manufacturer data
characteristics cannot be revealed due to non-disclosure agreements (NDAs) with participating
manufacturers. Thus, if the data used contained even a single manufacturer whose data required
a certain type of scaling, Table 5.28 lists that scaling was required for the entire aggregated
manufacturer data set. For details on the scaling procedures, see Section 5.4.6. Following Table
5.28, DOE presents the aggregated manufacturer results in Table 5.29,Table 5.30, Table 5.31,
and Table 5.32, for the 2.5W, 18W, 60W, and 120W representative units, respectively.
Table 5.28 Scaling and Markup Steps Performed On The Aggregated Manufacturer Data
Efficiency Efficiency
Cost
Cost
Rep.
Efficiency
Scaling
Scaling
Scaling Scaling Mark
Unit
Scaling
by
by
by
by
-Up to
Output
by Cord
Output
Output
Output
Cord
MSP
Power
Length
Power
Voltage
Power Length
2.5W

18W

60W

120W

Indicates that the data was scaled to the representative unit values, with effects on
efficiency or cost.
* Indicates that the data was scaled to the representative unit values, although the scaling had
no effect.

5-55

Table 5.29 Cost and Efficiency Relationship for 2.5W EPS (Manufacturer Interviews)
CSL 0
CSL 1
CSL 2
CSL 3
CSL 4
Mfr Unit Efficiency [%]:
58.3%
67.9%
71.0%
73.5%
74.0%
Mfr Unit No Load Power [W]:
0.500
0.300
0.130
0.100
0.053
Energy Star
Best in
Max
CSL Description:
EISA
Intermediate
2.0
Market
Tech
(58.3%, (67.9%,
(70.9%,
(73.2%, (73.9,
CSL Eff. [%], No-Load Power [W]
0.500)
0.300)
0.200)
0.100)
0.062)
MSP Difference with CSL0 [$]:
$0.00
$0.04
$0.23
$0.31
$0.42
Original Application:
N/A
N/A
N/A
N/A
N/A
Test Unit #:
N/A
N/A
N/A
996
N/A

Table 5.30 Cost and Efficiency Relationship for 18W EPS (Manufacturer Interviews)
CSL 0
CSL 1
CSL2
CSL 3
CSL 4
Mfr Unit Efficiency [%]:
76.0%
80.3%
83.0%
85.4%
87.5%
Mfr Unit No Load Power [W]:
0.500
0.300
0.200
0.100
0.062
Energy Star
Best in
Max
CSL Description:
EISA
Intermediate
2.0
Market
Tech
(76.0%, (80.3%,
(83.0%,
(85.0%, (87.3%,
CSL Eff. [%], No-Load Power [W]
0.500)
0.300)
0.200)
0.100)
0.062)
MSP Difference with CSL0 [$]:
$0.00
$0.32
$0.42
$0.79
$1.23
Original Application:
N/A
N/A
N/A
N/A
N/A
Test Unit #:
N/A
N/A
N/A
941
N/A

Table 5.31 Cost and Efficiency Relationship for 60W EPS (Manufacturer Interviews)
CSL 0
CSL 1
CSL2
CSL 3
CSL 4
Mfr Unit Efficiency [%]:
85.0%
87.0%
87.0%
88.0%
91.0%
Mfr Unit No Load Power [W]:
0.500
0.500
0.200
0.073
0.073
CSL Description:
EISA
Energy Star
Intermediate Best in
Max
2.0
Market
Tech
CSL Eff. [%], No-Load Power [W] (85.0%, (87.0%,
(87.0%,
(88.0%, (91.0%,
0.500)
0.500)
0.230)
0.210)
0.165)
MSP Difference with CSL0 [$]:
$0.00
$0.24
$0.71
$1.58
$2.90
Original Application:
N/A
N/A
N/A
N/A
N/A
Test Unit #:
N/A
N/A
N/A
838
N/A

5-56

Table 5.32 Cost and Efficiency Relationship for 120W EPS (Manufacturer Interviews)
CSL 0
CSL 1
CSL2
CSL 3
CSL 4
Mfr Unit Efficiency [%]:
85.0%
87.0%
88.0%
88.4%
91.7%
Mfr Unit No Load Power [W]:
0.500
0.500
0.230
0.210
0.165
CSL Description:
EISA
Energy Star
Intermediate Best in
Max
2.0
Market
Tech
CSL Eff. [%], No-Load Power [W] (85.0%, (87.0%,
(87.0%,
(88.0%, (91.0%,
0.500)
0.500)
0.230)
0.210)
0.165)
MSP Difference with CSL0 [$]:
$0.00
$0.66
$1.23
$1.41
$5.03
Original Application:
N/A
N/A
N/A
N/A
N/A
Test Unit #:
N/A
N/A
N/A
951
N/A
5.4.9

Summary of Results for Class A EPSs

Sections 5.4.7and 5.4.8 present engineering analysis cost-efficiency results that DOE
derived from test and teardown results as well as manufacturer interviews, respectively. Figure
5.29, Figure 5.30, Figure 5.31, Figure 5.32, Figure 5.33, Figure 5.34, Figure 5.35, and Figure
5.36 present results from both of these methods in terms of MSP difference versus average
efficiency and MSP difference versus no-load power consumption for each representative unit.
The label MSP Difference in these figures refers to the MSP above the baseline CSL0, as
opposed to the absolute MSP that a consumer would pay for the product. These graphs describe
the cost-efficiency relationship for EPSs, under the assumption that all other factors are held
constant. To that end, DOE normalized all representative unit data to the representative unit
criteria listed in Table 5.3.
In summary, for each representative unit, the manufacturer data for cost versus efficiency
and cost versus no-load power data showed a trend of increasing MSPs with higher CSLs,
meaning that it costs more to manufacture more efficient EPSs. DOEs SMEs agreed that, all
other factors being held constant, the cost-efficiency curves should be upwards sloping, whereby
more efficient EPSs correspond to higher MSPs. On the other hand, the testing and teardown
curves cost-efficiency relationship was inconclusive. In many cases for the test and teardown
curves, the CSL3 best-in-market unit was the cheapest, and the CSL0 unit was the most
expensive. Note that the testing and teardown results are presented only for CSL0, CSL1, and
CSL3 (CSL4 had no test unit to characterize it because it was the max-tech level, and CSL2 was
an intermediate level that was chosen after the testing and teardown analysis was complete).
For three of the four test and teardown representative units, the costs decreased between
CSL1 and CSL3. DOE believes that, although it rigorously normalized the test unit data, there
may have been factors that affect cost which were not normalized. In section 5.4.6,
manufacturers indicate that reliability is another factor that greatly impacts EPS cost and that it is
difficult to determine an EPSs reliability without access to its specification sheet. DOE did not
have access to EPS specifications and therefore could not control for reliability. Hence, DOE
believes the testing and teardown results to be inconclusive.
Although the trend of decreasing MSPs with higher efficiencies for Class A EPSs is
curious, DOE is aware of such instances for Non Class A EPSs. The cost-efficiency relationship
for high-power EPSs is an example of a negative-sloped curve, presented in the NCA EPS

5-57

Notice of Proposed Determination (NOPD) that DOE published on November 3, 2010. 74 FR


56928. DOE found that the efficiency-related materials cost of CSL0 was $115.32. The cost
decreased from $115.32 to $33.64 from CSL0 to CLS1. Then the cost increased from $33.64 to
$42.32 from CSL1 to CSL3. In this analysis, there is an underlying assumption that there is the
same utility between linear EPSs that characterize CSL0 and switched-mode EPSs that
characterize CSL1, CSL2, and CSL3. DOE did not receive any comments on the NOPD
disputing this assumption. Table II.29 and Figure II.7 in the NOPD provide further detail on the
cost-efficiency relationship of high-power EPSs. 74 FR 56928, 56954.
DOE plans to evaluate only the manufacturer data in its UEC, LCC, and NIA analyses for
Class A EPSs, as the testing and teardown data would lead to positive LCC results at CSL 3 for
all representative units where CSL 3 is the least expensive For the NOPR, DOE intends to
conduct another round of testing and teardowns for Class A EPSs that attempts to control for
additional factors. In particular, DOE intends to attempt to control the data for reliability by
selecting EPSs with similar end-use applications of similar quality. The results for the individual
representative units follow.

The 2.5W representative unit has manufacturer curves that are substantially different
from the testing and teardown curves. The manufacturer curves in both Figure 5.29 and Figure
5.30 are upwards sloping across increasing CSLs, with the greatest increase in MSP occurring
from CSL1 to CSL2. This is not the case with the testing and teardown curves. Figure 5.29 and
Figure 5.30 show an increase in MSPs from CSL0 to CSL1. However, the testing and teardown
data shows a decrease in cost from CSL1 to CSL2. Thus the testing and teardown data does not
indicate a clear trend. Therefore, DOE considers the testing and teardown data to be
inconclusive.
These inconclusive testing and teardown results might be due to not controlling for the
applications among the EPSs chosen, which can have many intricate effects, as manufacturers
suggested (see Section 5.4.5.6 for details). Specifically, as shown in Table 5.24, EPSs #876,
#935, and #996 are from cordless phone, generic, and mobile phone applications, respectively.
DOE expects that these applications require EPSs with different reliability specifications.

5-58

Manufacturer Selling Price Difference [$]

$0.50
$0.40
$0.30
$0.20
$0.10
Manufacturer Interview Data

$-

$(0.10)
Testing and Teardown Data

$(0.20)
$(0.30)
$(0.40)
57%

67%

77%

Average Active-Mode Efficiency [%]

$0.50
$0.40
$0.30

$0.20
$0.10
$0.00
-$0.10
-$0.20

-$0.30
-$0.40
600.00

400.00

200.00

Manufacturer Selling Price DIfference [$]

Figure 5.29 2.5W MSP vs. Efficiency curves for manufacturer and testing and teardown
data

Manufacturer Interview Data

Testing and Teardown Data

0.00

No-Load Power [mW]

Figure 5.30 2.5W MSP vs. No-load power curves for manufacturer and testing and
teardown data
The 18W representative unit has manufacturer curves that differ greatly from the testing
and teardown curves. The manufacturer curves in Figure 5.31 and Figure 5.32 are upwards
sloping, with the greatest increases in MSP occurring from CSL2 to CSL3 and CSL3 to CSL4.
The testing and teardown curves in both Figure 5.31 and Figure 5.32 are downwards sloping.

5-59

These downwards-sloping testing and teardown results might be due to not controlling
for the applications among the EPSs chosen, which can have many intricate effects. Specifically,
as shown in Table 5.25, EPSs #949, # 118, and #941 all have generic applications. Generic
EPSs can have significant variations in their specifications, as opposed to EPSs that are designed
for a specific product.

Manufacturer Selling Price Difference [$]

$1.50
$1.00

$0.50
$0.00
Manufacturer Interview Data

-$0.50
Testing and Teardown Data

-$1.00

-$1.50
75%

80%

85%

90%

Average Active-Mode Efficiency [%]

Figure 5.31 18W MSP vs. Efficiency curves for manufacturer and testing and teardown
data

$1.00
$0.50
$0.00
-$0.50

-$1.00
-$1.50

600

400

200

Manufacturer Selling Price Difference [$]

$1.50

Manufacturer Interview Data

Testing and Teardown Data

No-Load Power [mW]

Figure 5.32 18W MSP vs. No-load power curves for manufacturer and testing and
teardown data

5-60

The 60W representative unit has testing and teardown curves whose slope is steeper than
manufacturer curves between CSL0 and CSL1. The manufacturer curves in Figure 5.33 and
Figure 5.34 have the greatest increase in MSP from CSL3 to CSL4. Even though for the 60W
representative unit, the testing and teardown curves are upwards sloping, the mixed relationships
in the other three representative units leads DOE to consider this testing and teardown data
inconclusive as well.
One reason to suspect that even the 60W testing and teardown results are inconclusive is
that the analysis is not controlling for the applications among the EPSs chosen, which can have
many intricate effects, as manufacturers suggested (see Section 5.4.5.6 for details). Specifically,
as shown in Table 5.26, EPSs #999, # 834, and #838 are from notebook computer, generic, and
generic applications, respectively. Manufacturers indicated that generic EPSs have tremendous
variation in meeting an assortment of stringent specifications, as opposed to EPSs that are
designed with a specific product in mind.

Manufacturer Selling Price Difference [$]

The sharp rise in cost from CSL0 to CSL1 in Figure 5.34 results from manufacturers
only improving in the efficiency metric and not the no-load power metric from CSL0 to CSL1,
and hence the seemingly paradoxical increase in cost on the no-load power plot (Figure 5.34) is
explained by referring to the efficiency plot (Figure 5.33),

$3.50

$3.00
$2.50
$2.00
Manufacturer Interview Data

$1.50
$1.00

Testing and Teardown Data

$0.50
$0.00
84%

86%

88%

90%

92%

Average Active-Mode Efficiency [%]

Figure 5.33 60W MSP vs. Efficiency curves for manufacturer and testing and teardown
data

5-61

$3.00
$2.50
$2.00
$1.50

$1.00
$0.50
$0.00
600

400

200

Manufacturer Selling Price Difference [$]

$3.50

Manufacturer Interview Data

Testing and Teardown Data

No-Load Power [mW]

Figure 5.34 60W MSP vs. No-load power curves for manufacturer and testing and
teardown data
The 120W representative unit had manufacturer curves that looked very different from
the testing and teardown data. The manufacturer curves in both Figure 5.35 and Figure 5.36 are
non-decreasing across increasing CSLs, with the greatest increase in MSP occurring from CSL3
to CSL4. The testing and teardown curves, on the other hand, show a decrease in cost from
CSL0 to CSL1.
These inconclusive testing and teardown results might be due to not controlling for the
applications among the EPSs chosen, which can have many intricate effects, as manufacturers
suggested (see Section 5.4.5.6 for details). Specifically, as shown in Table 5.27, EPSs #854, and
# 951are both from notebook computers. Despite the same general application, manufacturers
indicated that another aspect of controlling for applications involved controlling for the quality
of the associated applications. For example, manufacturers indicated that not only is it necessary
to control for notebook computers, but it is also necessary to compare EPSs of brand-names with
applications of similar quality (i.e. low-end products to low-end products, and high-end products
to high-end products, etc.).
The sharp rise in cost from CSL0 to CSL1 in Figure 5.36 results from manufacturers
only improving in the efficiency metric and not the no-load power metric from CSL0 to CSL1,
and hence the seemingly paradoxical increase in cost on the no-load power plot (Figure 5.36) is
explained by referring to the efficiency plot (Figure 5.35),

5-62

Manufacturer Selling Price Difference [$]

$6.00
$4.00

$2.00
$0.00

Manufacturer Interview Data

-$2.00

Testing and Teardown Data

-$4.00
80%

-$6.00

85%

90%

95%

Average Active-Mode Efficiency [%]

Figure 5.35 120W MSP vs. Efficiency curves for manufacturer and testing and teardown
data

$4.00
$2.00
$0.00
-$2.00

-$4.00
600

400

200

No-Load Power [mW]

0
-$6.00

Manufacturer Selling Price Difference [$]

$6.00

Manufacturer Interview Data

Testing and Teardown Data

Figure 5.36 120W MSP vs. No-load power curves for manufacturer and testing and
teardown data

5.4.10 Creating Class A EPS CSL Equations for All Product Classes Based On the
Representative Units

5-63

DOE identified and selected four representative units within the representative product
class A1 on which to concentrate its analytical effort. DOE evaluated the representative units at
four CSLs based on four sources: the EISA 2007 standard, the Energy Star 2.0 specification, test
unit data, and manufacturer data. The following sections discuss how DOE scaled from
representative unit data to the CSLs of product class A1, and how DOE scaled A1s CSL
equations to product classes A2, A3, and A4, which DOE did not analyze directly.
5.4.10.1

Deriving CSL Equations for Product Class A1 From Representative Units

In its testing, DOE gathered data about individual Class A EPS representative units.
While these units provided insight pertaining to specific efficiency and no-load power values,
DOE needed to create CSL equations to account for all units. To that end, DOE created four
pairs of equations to define each CSL relating average efficiency to nameplate output power and
no-load mode power consumption to nameplate output power as specified in 5.4.2:
The equations for CSL0 are derived directly from the EISA standard.

The equations for CSL1 are derived directly from the Energy Star 2.0 Specification for

AC/DC basic voltage units.


The equations for CSL2 were created by curve-fitting the data points used to characterize
the Intermediate CSL, CSL2 (see section 5.4.2 for details on how the points were
selected).
The equations for CSL3 were created by curve-fitting best-in-market (BIM) test unit data
points.
The equations for CSL4 were created by curve-fitting data points from manufacturer
interviews.
The process of creating CSL2, CSL3 and CSL4 curves is described here in further detail.
DOE derived CSL2 by fitting equations to the efficiency values of the four Intermediate level
data points, at their respective output power values of 2.5W, 18W, 60W, and 120W (see
equations in Table 5.9). For all its curve-fitting, DOE used the simplest possible equation form
that was general enough to encompass all the CSL equations found in the EISA and Energy Star
2.0 CSLs. This equation was of the form Y=a*ln(Pout) + b*Pout+c, for each of the nameplate
output power segments, where Y indicates either the efficiency or no load-power requirement;
Pout indicates the nameplate output power; and a, b, and c indicate the specific parameters
defined in the respective CSLs.
Since, even this general form of the equation could not pass through all points exactly,
DOE ensured that the equations met three conditions. First, the distance to each point was
minimized. Second, the equation did not exceed the tested efficiencies. Third, DOE further
restricted the parameter choice in order to ensure that the CSL curves adhered to a matched pairs
approach. This means that the CSL curves yielded both non-decreasing (usually higher)
efficiency values and non-increasing (usually lower) no-load power values when moving from
lower to higher CSLs, across all output power levels. For the fitted CSL2 equation the maximum
difference between any data point and the equation was less than 0.5 percentage points.

5-64

The CSL2 no-load power equation was based on fitting the same equation forms on the
same nameplate output power segments as was done for CSL1, for the four Intermediate level
no-load power values at 2.5W, 18W, 60W, and 120W, without going below the chosen values
(see equations in Table 5.10). Specifically, Table 5.8 shows that CSL1 no-load power equations
had one constant value up to a nameplate output power of 50 watts, and another constant value
above 50 watts. Likewise, DOE based the CSL2 equations for no-load power in Table 5.10 on
the maximum no-load power among the two CSL2 lower power representative unit data points
(2.5W and 18W) and the two CSL2 higher power representative units data points (60W and
120W). The maximum of the lower power representative units was used for the first segment of
nameplate output power, of up to 50 watts, and the maximum of the higher power representative
units was used for the higher output power segment.
The CSL3 and CSL4 efficiency and no-load power specifications were created using the
same reasoning as CSL2. Instead of using the four Intermediate level data points, the CSL3
curves were fit to the BIM data points at 2.5W, 18W, 60W, and 120W (see equations for average
efficiency in Table 5.11 and no-load power in Table 5.12). Similarly, instead of using the four
Intermediate level data points, the CSL4 curves were fit to the CSL4 data points at 2.5W, 18W,
60W, and 120W, which were based on manufacturer interviews (see equations for average
efficiency in Table 5.13 and no-load power in Table 5.14).
5.4.10.2
A4)

Deriving CSL Equations for Non-Analyzed Product Classes (A2, A3, and

After developing the CSLs for product class A1, DOE developed additional CSL
equations for low voltage and AC/AC units. As noted in Section 5.4.2, Energy Star 2.0 has four
product classes, including low voltage and AC/AC product classes, each with its own set of
equations. DOE leveraged these existing Energy Star 2.0 equations in creating low voltage and
AC/AC CSLs.
For low voltage units, different CSL equations are necessary because low-voltage EPSs
have lower efficiency, as detailed in section 5.4.6.2. DOE believes that AC/AC units require
different CSL equations for no-load power based on Energy Star 2.0. Thus, there are four
possible combinations of CSL equations: the A1 CSL equations for basic voltage and AC/DC
no-load power (VBASIC and PNL_DC ), the low voltage equation for efficiency (VLOW), and the
AC/AC equation for no-load power (PNL_AC). Table 5.33 shows the efficiency equations, and
Table 5.34 shows the no-load power equations used for each of the four product classes.
Table 5.33 Active-Mode Efficiency CSL Equations for Product Classes A1, A2, A3, and A4.
Basic Voltage

Low Voltage

A1

A3

AC/AC

A2

Eff: VBASIC

AC/DC

Eff: VLOW
A4

Eff: VBASIC

Eff: VLOW

5-65

Table 5.34 No-Load Power CSL Equations for Product Classes A1, A2, A3, and A4.
Basic Voltage
A1

AC/DC
A3

AC/AC

Low Voltage

No load: PNL_DC

A2

No load: PNL_AC

A4

No load: PNL_DC
No load: PNL_AC

The VLOW efficiency equations were created using different methods, depending on the
CSL level. Though DOE had developed its own output voltage scaling method for scaling
representative unit data points, it was not used for product class scaling, because it was only
designed for the four specific representative units in product class A1 (see Section 5.4.6.2 for
additional reasons and details).
For CSL0, the curve is identical to the basic-voltage CSL0 curve because all units in the
market already meet the EISA standard. For CSL1, the VLOW equation is based directly on the
Energy Star 2.0 low-voltage equation.
The VLOW curves for CSL2, CSL3, and CSL4 were created by using their respective
CSL2, CSL3, and CSL4 VBASIC efficiency curves, and altering all equation parameters by the
difference in the coefficients between the CSL1 VBASIC and VLOW equations. This had the effect
of shifting the CSL2, CSL3, and CSL3 VLOW curves downwards from their corresponding VBASIC
CSL2, CSL3, and CSL4 curves, by a similar amount as the shift between the CSL1 VBASIC and
VLOW curves. The VLOW CSLs are shown the Table 5.35.

5-66

Table 5.35 The Low-Voltage Product Classes (A2, A4) Active Mode Efficiency Equations
CSL Nameplate Output Power (Pno) Minimum Average Efficiency in Active Mode
(expressed as a decimal)
CSL 0 Identical to Basic Voltage CSL 0
Identical to Basic Voltage CSL 0
0 to 1 watt
CSL 1

0.497 Pout + 0.067

> 1 to 49 watts

0.075 ln (Pout) + 0.561

> 49 watts

0.860

0 to 1 watt
CSL 2

0.507 Pout + 0.077

> 1 to 49 watts

0.0825 ln (Pout) - 0.0011 Pout + 0.586

> 49 watts

0.860

0 to 1 watt
CSL 3

0.517 Pout + 0.087

> 1 to 49 watts

0.0834 ln (Pout) - 0.0014 Pout + 0.609

> 49 watts

0.870

0 to 1 watt
CSL 4

0.537 Pout + 0.097

> 1 to 49 watts

0.0914 ln (Pout) - 0.0014 Pout + 0.609

> 49 watts

0.900

The equation for CSL0 PNL_AC is derived directly from the EISA standard, since all
EPSs meet this level. The equation for CSL1 PNL_AC is derived directly from the corresponding
Energy Star 2.0 Specification for AC/AC units. To create AC PNL curves for CSL2, CSL3, and
CSL4, DOE used the same approach as was adopted in the CSL0 and CSL1 PNL_AC curves, in
which a constant no-load power limit was set for both the CSL0 and CSL1 PNL_AC curves.
Specifically, the PNL_AC equations were derived by adjusting the corresponding PNL_DC curve
such that the AC PNL curve was constant across output power, by taking the highest of the
maximum allowable no-load power values across all nameplate output powers for the PNL_DC
curve. The PNL_AC CSLs are shown in Table 5.36.

5-67

Table 5.36 AC/AC Product Classes (A3, A4) Maximum Allowable No-Load Power
Equations
CSL Nameplate Output Power (Pno) Maximum Power in No-Load
Ac-Ac EPS
CSL 0
0 to 250watt
0.500 watts
CSL 1

0 to 250watt

0.500 watts

CSL 2

0 to 250watt

0.230 watts

CSL 3

0 to 250watt

0.210 watts

CSL 4

0 to 250watt

0.165 watts

5.5

NON-CLASS A EXTERNAL POWER SUPPLY ENGINEERING ANALYSIS

In this section DOE discusses how it intends to develop the NCA EPS engineering
analysis for the NOPR stage of the BCEPS rulemaking. First DOE describes which NCA EPS
representative product classes and representative units it will analyze. Next DOE discusses how
it will create CSLs for each type of NCA EPS. Then DOE explains why it believes that some of
the cost-efficiency analysis preformed for the Class A EPSs applies to some NCA EPSs. Finally,
DOE describes how it will apply to the NOPR the NCA EPS determination analysis that it
published on 14 May 2010. 75 FR 27170.
DOE based the NCA EPS engineering analysis on the analysis from the NCA EPS
determination analysis as well as the analysis developed in this preliminary analysis for Class A
EPSs. DOE based the analysis for multiple-voltage and high-power EPSs on the analysis
presented in the determination analysis because DOE believes the analysis to still be valid, in
part because interested parties did not question its validity. DOE based the analysis for medical
and MADB EPSs on the analysis presented in the preliminary analysis for Class A EPSs because
DOE believes there to be no technical limitations for medical and MADB EPSs to achieve the
same efficiencies at the same costs as Class A EPSs. Further, DOE presented this concept in the
determination analysis and interested parties did not question its validity, so long as MADB
EPSs do not include charge control.
5.5.1

Non-Class A EPS Representative Product Classes and Representative Units

In this section DOE presents its selection of NCA EPS representative product classes and
representative units based on the NCA EPS product classes presented in chapter 3. As noted
previously, DOE based the NCA EPS product classes on the product classes from the NCA EPS
determination analysis as well as the product classes developed in this preliminary analysis for
Class A EPSs.
5.5.1.1 Representative Product Classes and Units for Multiple-Voltage EPSs
DOE chose to treat both multiple-voltage EPS product classes as representative because
there are significant differences between the two classes as shown in Table 5.37. As explained in

5-68

detail in the determination analysis, representative product class X1 consists mainly of EPSs for
imaging equipment. These multiple-voltage EPS have similar nameplate output power ratings
for each output voltage. In contrast, EPSs in product class X2 are for video game applications
and typically have one output that provides the majority of the output power. Since a single
application constitutes the majority of shipments in each representative product class, DOE
elected to use those applications as representative units as shown in Table 5.38.
Table 5.37 Multiple Voltage EPS Representative Product Classes
Nameplate Output Power
Product Class
< 100 watts
X1 (representative
100 watts
X2 (representative)
Table 5.38 Multiple-Voltage EPS Representative Units
Nameplate Output
Nameplate Output
Second Nameplate
Power [W]
Voltage [V]
Output Voltage [V]
40
16
32
203
5
12

Example Application
Multi-Function Device
Video Game

5.5.1.2 Representative Product Classes and Units for High-Power EPSs


DOE chose to treat the high-power EPS product classes as representative because it is the
only product class (shown in Table 5.39). As explained in detail in the determination analysis,
representative product class H1 consists mainly of EPSs for ham radios. Since a single
application constitutes the majority of shipments in the representative product class, DOE elected
to use that application to define the attributes of the representative unit shown in Table 5.40.
Table 5.39 High Power EPS Representative Product Classes
Nameplate Output Power Product Class
> 250 watts
H1 (representative)
Table 5.40 High-Power EPS Representative Units
Nameplate Output Power [W] Nameplate Output Voltage [V] Example Application
345
13.8
Amateur Radio
5.5.1.3 Representative Product Classes and Units for Medical EPSs
DOE chose to treat only medical EPS product class M1 as representative because there
are significant similarities between the product classes shown in Table 5.41. Specifically, the
medical EPS product classes have the same product class structure as the Class A EPS product
classes, as explained in chapter 3. To that end, DOE applied the same structure for the medical
EPS representative product classes and representative units (shown in Table 5.42).
DOE took this approach because it is consistent with the approach for medical EPSs in
the determination analysis. Specifically, in the determination analysis DOE created
representative product classes and units for medical EPSs based on Class A EPSs. Interested
parties did not question the validity of this approach in the determination analysis. Since
publishing the determination analysis DOE further developed its understanding of Class A EPSs.
5-69

Consequently, the representative product classes and units presented in Table 5.41 and Table
5.42 are consistent with the Class A EPS analysis in section 5.4.
Table 5.41 Proposed Medical EPS Representative Product Classes
Basic Voltage Output Low Voltage Output*
AC-DC Conversion M1 (representative)
M2 (scaled)
AC-AC Conversion
M3 (scaled)
M4 (scaled)
* Low voltage output EPSs have nameplate output voltage less than six volts and nameplate output current greater
than or equal to 550 milliamps. All other EPSs are basic voltage output.

Table 5.42 Possible Medical EPS Representative Units


Representative Nameplate Output
Nameplate Output
Unit
Power
Voltage
[watts]
[volts]
1
2.5
5
2
18
12
3
60
15
4
120
19

Example
Application
Blood Pressure Monitor
Medical Nebulizer
Sleep Apnea Machine
Portable Oxygen Concentrator

5.5.1.4 Representative Product Classes and Units for MADB EPSs


DOE chose to treat only MADB EPS product class B1 as representative because there are
significant similarities between the product classes as shown in Table 5.43. Specifically, the
MADB EPS product classes have the same product class structure as the Class A EPS product
classes, as explained in chapter 3. To that end, DOE applied the same structure for the MADB
EPS representative product classes and representative units (shown in Table 5.44). Since DOE
can apply all Class A EPS analysis to MADB EPSs it intends to include a 120-watt
representative unit. However, Table 5.44 reflects that DOE has not identified an associated
application for this unit. As noted in the executive summary, DOE is seeking stakeholder
comment on any applications that use a 120-watt MADB EPS and whether to analyze this unit if
it cannot identify any associated applications.
DOE took this approach because it is consistent with the approach for MADB EPSs in
the determination analysis. Specifically, in the determination analysis DOE created
representative product classes and units for MADB EPSs based on Class A EPSs. Interested
parties did not question the validity of this approach in the determination analysis. Since
publishing the determination analysis DOE further developed its understanding of Class A EPSs.
Consequently, the representative product classes and units presented in Table 5.43 and Table
5.44 are consistent with the Class A EPS analysis in section 5.4.
Table 5.43 Proposed MADB EPS Representative Product Classes
Basic Voltage Output Low Voltage Output*
AC-DC Conversion B1 (representative)
B2 (scaled)
AC-AC Conversion
B3 (scaled)
B4 (scaled)
* Low voltage output EPSs have nameplate output voltage less than six volts and nameplate output current greater
than or equal to 550 milliamps. All other EPSs are basic voltage output.

5-70

Table 5.44 Possible MADB EPS Representative Units


Representative Nameplate Output Power Nameplate Output Voltage
Unit
[watts]
[volts]
1
2.5
5
2
18
12
3
60
15
4
120
19
5.5.2

Example
Application
RC Toy
DIY Power Tool
Motorized Bicycle
[none identified]

Non-Class A EPS Candidate Standard Levels of Efficiency

In this section DOE presents its selection of CSLs for NCA EPS and how it will develop
the CSL equations in the NOPR. DOE based the CSLs on the CSLs from the determination
analysis as well as CSLs developed in this preliminary analysis for Class A EPSs.
5.5.2.1 Candidate Standard Levels of Efficiency for Multiple-Voltage EPSs
DOE developed CSLs for multiple-voltage EPSs based on those presented in the NCA
EPS determination analysis. Specifically, multiple-voltage EPSs are distinct from other types of
EPSs both in that their underlying technology is different as well as that there are not established
standard levels in their market. Consequently, DOE structured the CSLs for multiple-voltage
EPSs based on products available in the market and the theoretical maximum technologically
feasible level described by manufacturers as shown in Table 5.45.
Table 5.45 Multiple-Voltage EPS Candidate Standard Levels of Efficiency
Number Reference
Basis
CSL0
Market Bottom Least efficient units in the market
CSL1
Mid Market
Efficiency in the middle of the market
CSL2
Best in Market Curve fit to most efficient test unit data points
CSL3
Max Tech
Curve fit to manufacturer max tech data points
In the determination analysis DOE defined the efficiency values for each CSL based on
test data and manufacturer data. For the NOPR, DOE intends to gather more data for multiplevoltage EPSs with which to develop specific efficiency values for each CSL. In contrast to Class
A EPSs whose minimum average efficiency and maximum no-load power consumption
requirements are functions of nameplate output power, DOE intends to have a single efficiency
metric for each CSL. DOE believes this approach is appropriate because although product
classes X1 and X2 span a range of nameplate output powers, each product class has only one
significant application and therefore only needs one value.
5.5.2.2 Candidate Standard Levels of Efficiency for High-Power EPSs
DOE developed CSLs for high-power EPSs based on those presented in the NCA EPS
determination analysis. Specifically, high-power EPSs are distinct from other types of EPSs both
in that their underlying technology is different as well as that there are not established standard
levels in their market. Consequently, DOE structured the CSLs for high-power EPSs based on

5-71

products available in the market and the theoretical maximum technologically feasible level
described by manufacturers as shown in Table 5.46.
Table 5.46 High-Power EPS Candidate Standard Levels of Efficiency
Number Reference
Basis
CSL0
Market Bottom Least efficient units in the market
CSL1
Mid Market
Efficiency in the middle of the market
CSL2
Best in Market Curve fit to most efficient test unit data points
CSL3
Max Tech
Curve fit to manufacturer max tech data points
In the determination analysis DOE defined the efficiency values for each CSL based on
test data and manufacturer data. For the NOPR, DOE intends to gather more data for high-power
EPSs with which to develop specific efficiency values for each CSL. In contrast to Class A
EPSs whose minimum average efficiency and maximum no-load power consumption
requirements are functions of nameplate output power, DOE intends to have a single efficiency
metric for each CSL. DOE believes this approach is appropriate because although product class
H1 spans a range of nameplate output powers, it has only one significant application and
therefore only needs one value.
5.5.2.3 Candidate Standard Levels of Efficiency for Medical EPSs
DOE developed CSLs for medical EPSs based on the CSLs for product class A1 for
Class A EPSs presented in section 5.4.2. Specifically, medical EPSs are similar to Class A EPSs
both in that their underlying technology is similar as well as that there are established standard
levels that can be applied to their market. However there are two differences between medical
EPSs and Class A EPSs, one of which affected DOEs choice of CSLs. First, medical EPSs are
exempt from EISA regulations so DOE expects that there are medical EPSs less efficient than
EISA standards. Second, as noted in chapter 3, medical EPSs conform to additional safety
requirements. These requirements make the absolute price of medical EPSs higher than Class A
EPSs, but do not affect medical EPS efficiency or incremental cost. Consequently, DOE
structured the CSLs for medical EPSs based on the Class A EPS CSLs and set the baseline CSL0
to the least efficient medical EPS units in the market as shown in Table 5.47 All other CSL
equations come directly from the Class A EPS analysis. For example, the previous baseline
CSL0 for Class A EPSs is CSL1 for medical EPSs.
Table 5.47 Medical EPS Candidate Standard Levels of Efficiency
Number Reference
Basis
CSL0
Market Bottom Least efficient units in the market
CSL1
EISA 2007
EISA 2007 equations for efficiency and no-load power
CSL2
Energy Star 2.0 Energy Star 2.0 equations for efficiency and no-load power
CSL3
Best in Market Curve fit to most efficient test unit data points
CSL4
Max Tech
Curve fit to manufacturer max tech data points
In the determination analysis DOE defined the efficiency values for each CSL based on
existing levels in the market, test data and manufacturer data. For the NOPR, DOE is
considering gathering more data for medical EPSs with which to develop specific efficiency

5-72

values the baseline CSL0. This would allow DOE to characterize the benefits of higher standards
to consumers purchasing EPSs with efficiencies below EISA requirements. However, DOE
believes that those consumers constitute a small and shrinking part of the market, based on
manufacturer interviews. Therefore, DOE is also considering accounting for the lowest part of
the market in the EISA qualified CSL as a simplifying alternative. Under this alternate
approach, there would be no difference between Class A EPSs and medical EPSs in the
engineering analysis. As noted in the executive summary, DOE is seeking stakeholder comment
on which approach to pursue.
Regardless of how DOE treats EPSs with efficiency below EISA, DOE intends to apply
data from Class A EPSs directly to medical EPSs for those CSLs that are common between the
two products. For each CSL, DOE intends to have equations that are functions of nameplate
output power similar to Class A EPSs. In particular, DOE intends to apply the Class A EPS
equations directly to the corresponding medical EPS CSLs. In addition, to scale medical EPS
CSL equations from product class M1 to the other product classes DOE intends to use the Class
A EPS methodology and results. DOE believes this approach is appropriate because of the
similarity between medical EPSs and Class A EPSs.
5.5.2.4 Candidate Standard Levels of Efficiency for MADB EPSs
DOE developed CSLs for MADB EPSs based on the CSLs for product class A1 for Class
A EPSs presented in section 5.4.2. Specifically, MADB EPSs are similar to Class A EPSs both in
that their underlying technology is similar as well as that there are established standard levels
that can be applied to their market. However there is one difference between MADB EPSs and
Class A EPSs, which affected DOEs choice of CSLs. Namely, MADB EPSs are exempt from
EISA regulations so DOE expects that there are medical EPSs less efficient than EISA standards.
In contrast to medical EPSs, DOE believes that MADB EPSs do not conform to additional
requirements, safety or otherwise. Consequently, DOE structured the CSLs for MADB EPSs
based on the Class A EPS CSLs and set the baseline CSL0 to the least efficient MADB EPS
units in the market as shown in Table 5.48. All other CSL equations come directly from the
Class A EPS analysis. For example, the previous baseline CSL0 for Class A EPSs is CSL1 for
MADB EPSs.
Table 5.48 MADB EPS Candidate Standard Levels of Efficiency
Number Reference
Basis
CSL0
Market Bottom Least efficient units in the market
CSL1
EISA 2007
EISA 2007 equations for efficiency and no-load power
CSL2
Energy Star 2.0 Energy Star 2.0 equations for efficiency and no-load power
CSL3
Best in Market Curve fit to most efficient test unit data points
CSL4
Max Tech
Curve fit to manufacturer max tech data points
In the determination analysis DOE defined the efficiency values for each CSL based on
existing levels in the market, test data and manufacturer data. For the NOPR, DOE is
considering gathering more data for MADB EPSs with which to develop specific efficiency
values the baseline CSL0. This would allow DOE to characterize the benefits of higher standards
to consumers purchasing EPSs with efficiencies below EISA requirements. However, DOE

5-73

believes that those consumers constitute a small and shrinking part of the market, based on
manufacturer interviews. Therefore, DOE is also considering accounting for the lowest part of
the market in the EISA qualified CSL as a simplifying alternative. Under this alternate
approach, there would be no difference between Class A EPSs and MADB EPSs in the
engineering analysis. As noted in the executive summary, DOE is seeking stakeholder comment
on which approach to pursue.
Regardless of how DOE treats EPSs with efficiency below EISA, DOE intends to apply
data from Class A EPSs directly to MADB EPSs for those CSLs that are common between the
two products. For each CSL, DOE intends to have equations that are functions of nameplate
output power similar to Class A EPSs. In particular, DOE intends to apply the Class A EPS
equations directly to the corresponding MADB EPS CSLs. In addition, to scale MADB EPS
CSL equations from product class M1 to the other product classes DOE intends to use the Class
A EPS methodology and results. DOE believes this approach is appropriate because of the
similarity between MADB EPSs and Class A EPSs.

5.5.3

Developing Cost-Efficiency Results for NCA EPSs

In this section DOE presents its methodology to develop cost-efficiency curves for NCA
EPSs. DOE intends to develop cost-efficiency curves based the data used to generate costefficiency curves in the determination analysis as well as the data used to generate costefficiency curves for Class A EPSs in this preliminary analysis.
To develop the NOPR, DOE intends to use the same scaling and analysis techniques for
both Class A EPSs and non-Class A EPSs. Thus, DOE intends to revise the data from the NCA
EPS determination analysis for the NOPR. Specifically, in developing the Class A EPS
engineering analysis in this preliminary analysis, DOE refined the scaling methodologies it used
in the NCA EPS engineering analysis in the determination analysis. In addition, DOE evaluated
EPSs based on the full costs of all materials, as opposed to the determination analysis that
evaluated EPSs based on the efficiency-related materials cost. DOE believes that its refined
scaling methods have enhanced robustness and that the full cost of an EPS is a more appropriate
basis for evaluation because of the inter-relatedness of EPS components. In the following
subsections DOE explains how it will use data from the determination analysis; in all cases, DOE
would apply the updated scaling and analysis techniques to the data.
5.5.3.1 Cost-Efficiency Curves for Multiple-Voltage EPSs
In the determination analysis DOE developed cost-efficiency curves for multiple-voltage
EPSs based on testing and tearing down EPSs as well as interviewing manufacturers; DOE
intends to follow a similar approach in the NOPR. Specifically, DOE believes that the costefficiency curves for multiple-voltage EPSs that it presented are accurate since it did not receive
comment from interested parties indicating otherwise. Therefore, DOE intends to use that data
(with updated scaling and analysis techniques) to develop the cost-efficiency curves in the
NOPR for product classes X1 and X2.

5-74

5.5.3.2 Cost-Efficiency Curves for High-Power EPSs


In the determination analysis DOE developed cost-efficiency curves for high-power EPSs
based on testing and tearing down EPSs as well as interviewing manufacturers; DOE intends to
follow a similar approach in the NOPR. Specifically, DOE believes that the cost-efficiency
curves for high-power EPSs that it presented are accurate since it did not receive comment from
interested parties indicating otherwise. Therefore, DOE intends to use that data (with updated
scaling and analysis techniques) to develop the cost-efficiency curves in the NOPR for product
class B1.
5.5.3.3 Cost-Efficiency Curves for Medical EPSs
To develop cost-efficiency curves for product class M1 for medical EPSs in the NOPR
DOE intends to apply directly the cost-efficiency curves for Class A EPSs. In addition, DOE
will conduct testing, teardowns, and manufacturer interviews to gather data on EPSs with
efficiency less than EISA.
5.5.3.4 Cost-Efficiency Curves for MADB EPSs
To develop cost-efficiency curves for product class B1 for MADB EPSs in the NOPR
DOE intends to apply directly the cost-efficiency curves for Class A EPSs. In addition, DOE
will conduct testing, teardowns, and manufacturer interviews to gather data on EPSs with
efficiency less than EISA.
5.6

BATTERY CHARGER ENGINEERING ANALYSIS


The battery charger engineering analysis estimates the cost associated with increasing the
efficiency of a representative BC. Due to the wide variety of BCs available in the market, DOE
analyzed only a limited number of representative product classes. These are presented in
section 5.6.1. A general discussion of the analytical methods used with each of the representative
product classes follows, including a discussion of the selection of representative units
(section 5.6.2), an evaluation of efficiencies in the market and development of CSLs
(section 5.6.3), and an evaluation of the associated costs through teardowns (section 5.6.4) and
manufacturer interviews (section 5.6.5). Finally, section 5.6.6 describes the markups applied to
the efficiency-related production costs to arrive at manufacturer selling prices for use in later
analyses, while section 5.6.7 describes how results from the analysis of the representative units
were extrapolated to the remainder of BCs within each representative product class.
Please see sections 5.6.8 through 5.6.13 for the application of the above methods to each
of the representative product classes, and section 5.6.14 for a discussion of extending this
analysis to the other BC product classes not analyzed directly.
5.6.1

BC Representative Product Classes


As discussed in the product classes section of the market and tech assessment (chapter 3),
DOE divided BCs into 10 product classes for the preliminary analysis based primarily on battery
voltage (in volts) and energy (in watt-hours). DOE divided BCs into additional classes based on
5-75

the type of input (AC or DC, inductive versus galvanic) and the type of output (AC output for
uninterruptible power supplies).
Although the divisions in voltage and energy between product classes were based on
differences in topology, i.e., the underlying design of the BC power converter, there are still
some similarities between the product classes that allow DOE to focus its analysis on a subset of
classes and apply the results to the remainder.
The classes that were explicitly analyzed through testing, teardowns, and interviews with
manufacturers are called representative product classes. These were chosen as the representative
product classes because they collectively constitute the majority of BC shipments and national
energy consumption. By focusing its analysis on the representative product classes, DOE can
provide the most accurate standards analysis for the applications that have the largest impact on
individual users (due to high energy consumption) or that have the largest impact on the nation
(due to high shipment-weighted energy consumption).
To select the representative product classes from among the ten product classes outlined
in chapter 3, DOE calculated the energy consumption of the applications found in the market
survey, factoring in typical shipments, lifetimes, operating hours, and efficiencies for BCs across
a wide range of applications. The distribution of annual shipments by product class is presented
in Figure 5.37 while the distribution of annual energy consumptionc is presented in Figure 5.38.
The representative product classes are also summarized in Table 5.49.

Annual energy consumption and unit energy consumption are used interchangeably in this document.

5-76

Figure 5.37. Distribution of annual shipments by product class.

Figure 5.38. Distribution of annual energy consumption by product class.

5-77

Table 5.49 Representative Product Classes


#

<
100
Wh
AC In,
DC Out
100
3000
Wh

Inductive
Connection

0-4 V

410 V

1048 V

024 V

2448 V

> 3000 Wh
DC In,
DC Out
AC In,
AC Out

5 V Input

12 V Input

AC Output
from
Battery

10

Product Class
Low Energy,
Inductive
Low Energy,
Low Voltage
Low Energy,
Med. Voltage
Low Energy,
High Voltage
Med. Energy,
Low Voltage
Med. Energy,
High Voltage
High Energy
Low Energy,
5V DC Input
Low Energy,
12V DC Input
Low Energy,
AC Output

Representative Product Class


or Scaled Product Class
Representative Product Class
Representative Product Class
Scaled
Representative Product Class
Representative Product Class
Scaled
Representative Product Class
Scaled
Scaled
Representative Product Class

5.6.2

BC Representative Units
DOE focused its engineering analysis for each BC representative product class on one
representative unit, an idealized BC typical of those found in the product class. Because results
from the analysis of these representative units would later be extended to additional BCs, DOE
selected them from high-volume and/or high-energy-consumption applications, as determined by
the market survey. Nonetheless, the analysis of these BCs is pertinent to all the applications in
the product class under the aforementioned assumption that all BCs with the same battery
voltage and energy provide similar utility to the user, regardless of the actual end-use product
they work with.
DOE evaluated the data from the market survey and, for each representative product
class, identified the most common battery voltage and energy combinations. DOE then selected
the representative units to correspond to the most common combinations of battery voltage and
energy that also incorporated a wide variety of applications. By selecting representative unit
characteristics (battery voltage and energy) common to BCs for several applications, DOE (1)
extended the applicability of the analysis across a larger portion of BCs and (2) increased the
variety of efficiencies used in its analysis.
To elaborate, the primary benefit of focusing on battery voltage and energy pairings that
were typical of popular BC products across a variety of applications was the wider applicability
5-78

of the resultant analysis. By treating the BC component of multiple applications as an


interchangeable component and analyzing not onlyfor examplecellular telephone BCs, but
also those for cordless telephones and digital cameras, DOE ensures that the BC representative
unit is representative of all applications for BCs with the representative-unit battery voltage and
energy.
As an added benefit, evaluating BCs for multiple applications will also result in a greater
variety of costs and efficiencies represented in the analysis. Just as the end-use product
applications vary in cost and size, so do their BCs. Now, because these two characteristics of a
BC impact its efficiency,d the efficiency of BCs tends to stratify by application. By analyzing
multiple applications, DOE therefore ensures that its analysis takes into account the full variety
of BCs efficiencies in the market.
Figure 5.39 shows the characteristics of the BC representative units superimposed over
the results of the market survey, previously presented in Figure 5.3 The figure illustrates how
the representative units compare to the data obtained through the market surveys. Additional
information regarding the representative units is compiled in Table 5.50. In addition to the
battery energy and voltage, the table also displays typical production volumes for a single BC
model in each product class. These volumes were used only in the teardown portion of the
engineering analysis, where they were used to calculate how the fixed costs of manufacturing
equipment and non-recurring engineering costs would be amortized over the total number of BCs
in a production run, and their impact on the unit price.

Efficiency has a real impact on the minimum size of electronic components, as the components ability to dissipate
heat is constrained by their available surface area. The smaller the component, the less surface area it has. Therefore,
requirements on size and weight often motivate more efficient designs.

5-79

Table 5.50 The BC representative units for each representative product class.
Annual
Lifespan
Rep.
Battery Battery
Production
Production
Unit
Product
Voltage Energy
Volume
Volume
Typical
#
Class
V
Wh
K units
K units
Applications
Low Energy,
3.6
1.5
500
2000
Toothbrushes
1
Inductive
Low Energy,
Cellular and
3.6
3
480
1600
2
Low Voltage
Cordless Phones
Low Energy,
Scaled Product Class
3
Med. Voltage
Notebook
Low Energy,
10.8
20
640
2180
Computers,
4
High Voltage
Power Tools
Marine
Med. Energy,
12
800
50
500
Chargers,
5
Low Voltage
Wheelchairs
Med. Energy,
Scaled Product Class
6
High Voltage
7
8
9
10

High Energy
Low Energy,
5V DC Input
Low Energy,
12V DC
Input
Low Energy,
AC Output

48

3750

150

1460

Golf Carts

Scaled Product Class


Scaled Product Class
12

70

1000

5-80

5000

Uninterruptible
Power Supplies

Figure 5.39 The BC representative units superimposed over the BC market survey results.

5-81

5.6.3

BC Candidate Standard Levels of Efficiency


Subsequently, DOE examined the impacts on cost of improving the efficiency of each of
the representative units presented in section 5.6.2 to evaluate the impact and assess the viability
of potential energy efficiency standards. As described in the technology assessment and
screening analysis, chapters 3 and 4, the technology options for improving efficiency are many;
each incremental technology improvement increases the BC efficiency along a continuum. The
engineering analysis develops cost estimates for several CSLs along that continuum.
CSLs are often based on (1) efficiencies available in the market; (2) other voluntary
specifications or mandatory standards that cause manufacturers to develop products at particular
efficiency levels; and (3) the maximum technologically feasible level.e
There are no current energy conservation standards for BCs, and the ENERGY STAR
efficiency1 level may not be widely applicable.f Therefore, DOE based the CSLs for its BC
analysis on the efficiencies attainable through the design options presented previously, as seen in
commercially available units. DOE selected commercially available BCs at the representativeunit battery voltage and energy from the high-volume applications identified in the market
survey. DOE then tested these in accordance with the proposed DOE BC test procedure. For
each representative unit, DOE then selected CSLs to correspond to the efficiency of BC models
that were comparable to each other in most respects, but differed significantly in efficiency.
In general, for each representative unit, DOE chose the baseline (CSL 0) unit to be the
one with the highest annual energy consumption, and the best-in-market (CSL 2) to be one with
the lowest. Where possible, the energy consumption of an intermediate model was selected as the
basis for CSL 1, to provide additional resolution to the analysis.g
Unlike the previous three CSLs, CSL 3 was not based on an evaluation of the efficiency
of BC units in the market, since BCs with maximum technologically feasible efficiency are not
commercially available due to their high cost. Where possible, Navigant Consulting, obtained
manufacturer estimates of max-tech costs and efficiencies. In some cases manufacturers were
unable to offer any insight into efficiencies beyond the best currently available in the market. An
independent estimate of the efficiency of a max-tech unit was also performed by DOE through
extrapolation from its analysis of the best-in-market (CSL 2) unit by estimating the impacts of
adding any remaining energy efficiency design options. When neither methods was feasible,
there is no data provided at this time for CSL 3.

The max-tech level represents the most efficient design that is commercialized or has been demonstrated in a
prototype with materials or technologies available today. Max-tech is not constrained by economic justification,
and typically is the most expensive design option considered in the engineering analysis.
f
The ENERGY STAR level for BCs was not adopted as a CSL because the ENERGY STAR BC guidelines are
limited in scope mainly to BCs for motor-operated applications and furthermore do not consider energy
consumption in active mode.
g
An alternative approach would have set CSL 1 at the best-in-market unit. The approach would have been used in
the absence of an intermediate, improved-efficiency unit. However, DOE decided against this approach to ensure
naming consistency.

5-82

The CSLs in Table 5-51 are presented in terms of annual energy consumption, with
higher CSLs corresponding to lower energy consumption, assuming a class-average usage
profile. However, to minimize the sensitivity of the CSL units to the particular usage profile
selected for a given product class, DOE based its CSLs on the successive BC units that show
efficiency improvement in all of the modesi.e., active, maintenance, no-battery, and off
modes. Thus, although the results of the analysis may change depending on the usage profile
selected, the ordering of the CSLs will not. The BC model corresponding to the baseline CSL
will continue to consume more energy than that corresponding to the next higher CSL, and so on,
regardless of the usage profile. These CSLs are summarized for each representative unit in Table
5-51, below, while the usage profiles used to calculate energy consumption at each CSL are
shown in Table 5.52.
Table 5-51. Energy Consumption of Battery Charger Representative Units at Each
Efficiency Level
Annual Energy Consumption by Representative Unit
kWh

# Efficiency Level
0
Baseline

10

Low
Energy,
Inductive

Low
Energy,
Low
Voltage

Low
Energy,
High
Voltage

Med.
Energy,
Low
Voltage

High
Energy

Low
Energy,
AC
Output

1.5 Wh,
3.6 V
10.0

3 Wh,
3.6 V
10.0

20 Wh,
10.8 V
40.9

800 Wh,
12 V
112.7

3750 Wh,
48 V
290.0

70 Wh,
12 V
65.2

7.0
5.1
10.9
78.7
250.0
19.6
1
Improved
3.5
1.2
6.4
45.0
200.0
6.4
2 Best-in-Market
10.0
150.0
1.5
3
Max-Tech*
* For some product classes, manufacturer interviews did not yield any cost-efficiency data at the
max-tech CSL.

5-83

Table 5.52 Weekly Usage Assumptions for Battery Charger Representative Units
Time Per Day Spent in Each Mode

1
2
4
5

Representative
Unit
Low Energy,
Inductive
Low Energy,
Low Voltage
Low Energy,
High Voltage
Med Energy,
Low Voltage

Active
and
Maint.
hr

Off
hr

Total
hr

Number
of Full
Charges
Per Day

NoBattery
hr

23.9

0.1

0.0

0.0

24.0

0.3

10.0

4.9

9.1

0.0

24.0

0.6

17.4

0.3

6.1

0.2

24.0

1.0

7.7

0.5

15.8

0.0

24.0

0.6

Unplugged
hr

High Energy

7.7

0.1

16.2

0.0

24.0

0.4

10

Low Energy,
AC Output

24.0

0.0

0.0

0.0

24.0

0.0

5.6.4

BC Teardown Analysis
As mentioned in the discussion above, the CSLs used in the BC engineering analysis
were based on the efficiencies of BCs available in the market. Following testing, the units
corresponding to each commercially available CSL were disassembled to (1) evaluate the
presence of energy efficiency design options and (2) estimate the materials cost. The
disassemblies were performed by one of DOEs subject-matter experts and included an
examination of the general design of the BC as well as an evaluation of the presence of any of
the technology options discussed in chapter 3.
After the BC units corresponding to the CSLs were evaluated, they were torn down by
iSuppli, another DOE contractor. For most BCs, the teardowns were comparable to those
conducted for EPSs, described in section 5.4.4.
Teardowns were done differently for BCs embedded inside complex consumer electronic
products such as camcorders and notebook computers. Because the BC constitutes a small
portion of the circuitry of these products, DOE did not evaluate the entirety of the products cost.
Rather, iSuppli identified the subset of components in each product enclosure responsible for
battery charging, including the battery, charge regulator, and any related power converters and
voltage regulators.
In general, any component in the product and enclosure can be categorized as (1)
intended solely for battery charging functions, such as the battery itself; (2) intended solely for
non-battery charging functions, such as a user-interface component; and (3) intended for both
battery charging and non-battery charging functions; such as a power supply. For the preliminary
analysis, iSuppli included in the BC bill-of-materials components in categories 1 and 3, the latter
because of their crucial role in the battery charging process. Nonetheless, this choice was not

5-84

always appropriate and cases where dual-purpose components such as microcontrollers


unnecessarily inflated the cost of the BC are noted in the subsequent sections.
5.6.5

BC Manufacturer Interviews
In January, February, and March of 2010, Navigant Consulting, Inc. (Navigant
Consulting) interviewed several battery charger (BC) manufacturers on behalf of the U.S.
Department of Energy (DOE). Of these, some manufactured the BCs directly, while others were
original equipment manufacturers (OEMs) of battery-operated products. The purpose was to
obtain data on the possible efficiencies and resultant costs of consumer BCs. Navigant
Consulting entered into non-disclosure agreements whereby it can present to DOE general
information about the EPS market and technology, but no confidential data specific to any
individual manufacturer.
Prior to the interview, each manufacturer was sent a Cost-Efficiency Estimation Survey
(included as appendix 5C) to guide responses. To ensure consistency between manufacturers,
the survey specified the parameters of each BC representative unit under consideration,
previously presented in section 5.6.2.
5.6.5.1

Manufacturer Responses

For each representative unit, the interviewers asked manufacturers to describe the
technological improvements and associated costs necessary to meet each of the CSLs presented
in section 5.6.3. These CSLs were also presented in a disaggregated form (i.e., energy
consumption by mode, rather than combined into a weighted annual energy consumption) to help
the respondents.
Nonetheless, in many cases, manufacturers were unable to provide information in terms
of the representative unit, and responded instead with information on other popular models with
which they were more familiar. The output characteristics of these models are plotted in relation
to the representative units in Figure 5.40.

5-85

Power
Tools

Notebook
Computers

Figure 5.40 Representative units and units described by manufacturers.


The most significant differences between manufacturers units and the representative
units can be seen in the low energy, high voltage product class (20 Wh, 10.8V). This product
class includes notebook computers and power tools. In general, the notebooks align with the
representative unit battery voltage, but have higher energy, while the power tools align with the
battery energy, but have higher voltage. This is due to the competing requirements of the two
applications: batteries for notebooks are designed for maximum energy storage, while those for
power tools deliver maximum power. This leads to not only higher voltages in power tools, but
also different battery design. Some manufacturers felt that units within the same product class
can only be compared if they have the same chemistry (i.e., lithium-ion or nickel) and cell type
(i.e., power vs. energy cells).
Nonetheless, this will become less of a concern as products continue to shift from nickel
to lithium-ion batteries. For existing products, manufacturers reported that despite different
OEMs and suppliers, there is similarity in design between chargers for various applications and
chemistries, with the same architectures and methods of operation. Furthermore, several
manufacturers revealed that the difference in charge rate between BCs for the two applications
a function of the intrinsic qualities of the cells (i.e., power vs. energy) as well as the usageis
shrinking, with both startups and established companies offering notebooks with charge rates
above 0.5 C (compared to 1 C for professional power tools). DOE welcomes further comment on
any dissimilarities between BCs for these two types of low energy, high voltage products, and
how best to address them in the analysis, e.g., through the creation of additional product classes.
Other discrepancies between the representative units and the manufacturers popular units
do not seem as significant. In general, manufacturers did not expect the energy consumption of a
BC to vary due to small variations in battery voltage and energyespecially at the higher battery
energies where there is a diminishing impact of fixed losses at higher output power. This same

5-86

effect is responsible for the flattening of the EPS efficiency curve at higher powers, discussed in
greater depth in section 5.4.6.1.
5.6.5.2 Aggregation of Manufacturer Responses
After collecting information from manufacturers on the mode-specific energy
consumption and cost of battery chargers, Navigant Consulting used the following process to
calculate aggregate manufacturer cost-efficiency data at each CSL:
1. Calculate the unit energy consumption for each manufacturer design;
basis function,
2. Perform a regression analysis of each manufacturers data using a
resulting in a cost-efficiency curve for each manufacturer defined for every UECh;
3. Translate each manufacturer curve to obtain incremental costs from the baseline;
4. Average all the manufacturers incremental costs at each CSL, resulting in an
incremental aggregate cost-efficiency curve;
5. Translate the aggregate curve to equal the teardown results at the baseline UEC,
resulting in an absolute incremental cost-efficiency curve; and
6. Finally, decompose the aggregated UECs at each CSL to estimate performance in
each mode for a typical manufacturer unit (used for comparison with teardown results
and for calculating application-specific energy consumption used in the LCC
analysis).
Each of these steps is described in the sections that follow.
Calculate the UEC for Each Manufacturer-Supplied Design
Navigant Consulting used the shipment-weighted average usage profile for each product
class to calculate the annual energy consumption of each BC when used with the typical mix of
applications for the product class. The derivation of the application-specific usage profiles, the
resultant product class-average usage profiles, and the calculation of the UEC are all described in
Chapter 7 of the TSD; nonetheless, these procedures were used within the engineering analysis
prior to reporting manufacturers cost-efficiency data as a necessary step in the aggregation
procedure.
Regression Analysis to Calculate Manufacturer Costs Across all UECs
However, the resulting shipment-weighted UECs did not necessarily coincide with the
UECs specified at each CSL; therefore, it was not possible to simply average the costs provided
by the manufacturers at each CSL to obtain an aggregate cost-efficiency curve. Instead, Navigant
Consulting performed a regression analysis of the manufacturer cost-efficiency data, to obtain a
best-fit cost-efficiency curve for each manufacturer.

Unit energy consumption (UEC) and annual energy consumption are used interchangeably in this document.

5-87

According to the manufacturer interviews, large improvements in energy efficiency could


initially be made at relatively little cost. For example, manufacturers reported that improved
transformer steel could decrease the energy consumption of low-energy, low-voltage chargers by
20 percent for a $0.25 increase in selling price. However, further improvements would offer
diminishing returns until the maximum-technologically achievable efficiency, beyond which no
further improvements could be possible given currently available technology. An ideal charger
with zero losses in active, maintenance, or no-battery modes, but with a lossy battery, could be
assumed to have an infinite selling price.
Navigant Consulting modeled this apparent relationship between cost and efficiency as a
1 curve of the form presented in Eq. 1, below:
x

MSP

1
m UEC b

a,

Eq. 1
Where:
MSP is the manufacturers selling price;

UEC is the unit energy consumption;

m is a parameter controlling the concavity of the curve;

a is a parameter corresponding to the flatness of the curve at the lower efficiencies; and

b is a parameter corresponding to the energy consumption of the ideal charger.

Curves of this form were fitted to the MSP-versus-shipment-weighted-UEC data for each
manufacturer; the a variable, above, was adjusted as necessary to maximize the coefficient of
determination (R2 value). This process is illustrated in Figure 5.41.

5-88

Figure 5.41 Illustration of the regression analysis of manufacturer cost-efficiency data.


Translate Curves to Obtain Incremental Costs from the Baseline
After calculating the manufacturer best-fit-curves, Navigant Consulting translated each
one by subtracting the cost at the baseline CSL. As illustrated in Figure 5.42, the cost translation
process shifts the absolute system costs of each manufacturers products so that the cost of the
baseline system is consistent across all manufacturers. By ensuring that all manufacturers share a
common baseline system cost, DOE was able to compare manufacturers data points directly
while maintaining each manufacturers incremental costs as efficiency increases, allowing for a
direct analysis of the cost of efficiency improvements regardless of baseline costs (which may
vary depending on the size of the manufacturer).

5-89

Figure 5.42 Illustration of translation of curves to obtain incremental costs from the
baseline
Average the Incremental Costs at Each CSL
The last step of the procedure produced cost-efficiency relationships with a common
baseline for each manufacturer across all UECs. Next, DOE aggregated these responses by
calculating the cost for each manufacturer at each CSL and averaging them, such that the
baseline aggregate cost was the average of the manufacturer costs at the baseline CSL 0 (equal to
zero from the last step), the aggregate cost at CSL 1 was the average of the manufacturer costs at
CSL 1, and so on. This process is illustrated in Figure 5.43.

Figure 5.43 Aggregation of translated manufacturer curves.

5-90

Translate Curves to Obtain Absolute Costs from Teardown Baseline


Finally, to allow easier comparison between manufacturer and teardown results and more
meaningful interpretation of analysis results as a portion of total unit costs, Navigant Consulting
translated the aggregate manufacturer curves such that the manufacturers MSPs at baseline
efficiency corresponded to the lowest teardown MSP. This process is illustrated in Figure 5.44.

Figure 5.44 Translation of curves to obtain absolute costs from teardown baseline
DOE realizes that other methods of normalizing the data are available, such as shifting
the cost values to center the curves at a market-weighted average or lowest-cost manufacturer
MSP, and such methods could be considered for use in further analyses.
Decompose the aggregate UECs at each CSL to Estimate Mode-Specific Performance
Finally, DOE decomposed the aggregate UEC at each CSL into mode-specific
performance figures. As an illustration, a UEC of 10 kilowatt-hours per year for a given
representative unit was decomposed into an estimated active-mode efficiency of 50 percent,
maintenance mode power of 2 watts, and no-battery mode power of 1 watt. These mode-specific
estimates were calculated taking into account the original manufacturer responses, which were
also provided in terms of mode-specific performance, and were used to provide a more direct
comparison to DOE test results and for use in the downstream LCC analysis.
Rather than weighting the original manufacturer responses to calculate the UEC,
followed by decomposing the aggregate UEC into mode-specific performance, Navigant
Consulting could have also aggregated the mode-specific responses directly. However, doing so
would have treated the performance in each mode as independent; by first calculating the UEC,
DOE takes into account the fact that manufacturers will tend to optimize performance in one
mode over that in another, depending on the usage of the product.
A summary of the aggregated manufacturer responses, divided by representative unit, is
presented in sections 5.6.8 through 5.6.13.
5-91

5.6.6

BC Cost Model

DOE gathered inputs on markups for BCs from manufacturer interviews. Specifically,
DOE questioned manufacturers regarding typical markups for a BC manufacturer, or original
device manufacturer (ODM), between bill-of-materials (BOM) cost, i.e., what they pay for the
BC components, and manufacturer selling price (MSP), i.e., the price at which they sell the BC
to an original equipment manufacturer (OEM). This MSP is alternatively known as the OEMs
assembly price or factory price.
The analysis focused on the cost of BC components directly related to efficiencyi.e.,
the electronics, and excluded the packaging, cord, and cosmetic touches which may vary from
product and depends greatly on the application, and whether or not the BC is integrated into a
product or packaged separately. Therefore, the resultant MSP was an electronics MSP.
In addition to ignoring packaging costs, DOE further simplified the analysis by
standardizing how BC production and non-production markups would be calculated across the
ODM and OEM, both of which may participate in the manufacturing of the BC. As further
explanation, the four possible arrangements for BC production are illustrated in Figure 5.45,
below.
1. The OEM manufactures both the end-use product and all BC components
2. The OEM manufactures the end-use product but purchases all BC components from an
ODM or ODMs
3. The OEM manufactures the end-use product, including any BC components embedded in
the end-use product, while purchasing remaining BC components from an ODM or
ODMs
4. The OEM manufactures nothing; instead it purchases the end-use product and all BC
components from an ODM or ODMs

5-92

Figure 5.45 The four ways that BC and end-use product manufacture can be divided
between the BC supplier or original device manufacturer (ODM) and the original
equipment manufacturer (OEM).
Which of the above arrangements applies to a particular BC will influence how the
various production and non-production markups are calculated. The behavior of the OEM can
further influence the markup. Some OEMs indicated that they pass through the cost of the BC,
while others claimed to mark it up according to the same schedule as the rest of the batteryoperated product.
Rather than attempt to take into account all the manufacturing relationships revealed
during the interviews, DOEs preliminary analysis assumes a standard BC manufacturing
markup, regardless of who manufactures the BCwhether it is the OEM or ODM. This markup
is an average of those provided by manufacturers and accounts for the production and nonproduction costs, as well as profit, associated with electronics assembly.
Although this markup, presented in Table 5.53, varies by product class, it is independent
of application, as the same factories typically manufacture BCs for a host of different
applications. The large variations in the markup can be explained by the nature of the product
sold. In general, the lower-energy units are manufactured in larger numbers (see Table 5.50) and
rely on somewhat generic EPSs for power conversion. In contrast, the higher-energy units are
manufactured in smaller numbers, and the same manufacturer typically manufactures the power
conversion and higher-value charge-control portions of the BC. DOE multiplied these productclass specific markups by the BOM cost to arrive at an MSP for each unit.

5-93

Table 5.53 Average markup applied to the BOM costs obtained both through teardowns as
well as manufacturer interviews to obtain the MSP.
Average BC
Representative
Manufacturer
Product Class
Markup
Low Energy,
1.2
1
Inductive
Low Energy,
1.2
2
Low Voltage
Low Energy,
1.5
4
High Voltage
Med Energy,
2
5
Low Voltage
7

High Energy

10

Low Energy,
AC Output

1.6

5.6.7

BC Scaling from Representative Units to Product Classes


Following the development of engineering results for the representative units, DOE must
extend its analysis to all BCs that were not specifically analyzed, but which would also be
covered by the standard. This task is twofold: (1) scaling the representative unit results to BCs
that are also in the representative product class, but which differ in battery voltage and energy;
and (2) scaling the representative unit results to BCs in product classes not explicitly analyzed.
Specific methodologies for scaling within each class are discussed in the section
pertaining to that class (sections 5.6.8 through 5.6.13, below), while scaling to the remainder of
the product classes is discussed in section 5.6.14.
There are nonetheless several general strategies that may be applicable to all product
classes, and could be used when no specific scaling can be developed for a particular product
class. These are: (1) using the scaling procedures developed for EPSs; (2) using scaling
relationships obtained through manufacturer interviews; or (3) using scaling relationships
obtained from analyzing BC test results. The three are introduced in the subsections that follow;
DOE seeks comment on the validity of each of these methods for calculating the energy
consumption and costs of other BCs following the analysis of the representative units.
5.6.7.1 Scaling Based on Procedures Developed for the EPS Analysis
As mentioned previously in 5.4.6, DOE developed scaling relationships that estimate how
the cost and efficiency of an EPS is likely to vary if redesigned for another output voltage and
power. The output voltage scaling relationships were developed by analyzing the results of
efficiency tests conducted on EPS familiesgroups of EPS made by the same manufacturer that
share the same design and output power but vary in output voltage. The output power scaling
relationships, on the other hand, are based on the established levels for EISA 2007 energy
conservation standards and ENERGY STAR 2.0 voluntary guidelines for Class A EPSs, which

5-94

were developed by examining the market and noting the distribution of efficiencies at each
output power. At output powers where the average efficiency distribution does not vary
significantly, the standard levels are flat, and there can be assumed to be no relation between
changes in output power and changes in achievable efficiency. The cost of an EPS was assumed
not to vary with voltage, but in cases of varying output power, DOE scaled the costs
proportionally with output power, as suggested by manufacturers.
Because of similarities between the design and construction of EPS and BC power
conversion circuits, those same relationships can be used to scale BC results. In particular, EPS
output power would be treated as analogous to battery energy and EPS output voltage as
analogous to battery voltage, while their impacts on EPS efficiency would translate to impacts on
BC energy consumption in active mode. Since energy consumption in no-battery and
maintenance modes is primarily dependent on overhead losses,i the energy consumption in these
two modes would not be scaled, consistent with the lack of scaling in no-load mode for EPSs.
5.6.7.2 Scaling Based on BC Manufacturer Interviews
During the interviews manufacturers relayed several methods for scaling the energy
consumption of a BC within a product class, and explained how certain circuit losses vary with
battery voltage and energy. Specifically, when asked how changes in output power are likely to
affect active-mode efficiency, some manufacturers stated that the active-mode efficiency of BCs
will tend to increase with output power, up to a point, while one manufacturer specifically
suggested using the curve contained in the ENERGY STAR specification for EPSs, as discussed
above.
Additionally, manufacturers also discussed non-active losses. Although some
manufacturers indicated that losses in no-battery and maintenance modes are both mostly
independent of the battery, others expected that they would vary with battery voltage. In
particular:
Standby or no-battery mode losses: A higher battery voltage would lead to higher BC
output power, requiring a larger transformer with consequently higher magnetization
losses;
Maintenance mode losses: Overhead losses would increase as above, plus the
maintenance power would increase due to the higher battery voltage (this power is the
product of maintenance current and battery voltage).
DOE could aggregate these and other manufacturer suggestions to develop general
methods for scaling energy consumption in each of the modes.
5.6.7.3 Scaling Based on Analysis of BC Test Results
Finally, rather than use the above two strategies (using scaling relationships derived from
the analysis of EPS results, or obtained from manufacturers), DOE could also develop novel
i

Except for slow, nickel-chemistry trickle chargers, which do not terminate their charge and therefore output
significant current in maintenance mode. In that case, maintenance mode would be treated similarly to active.

5-95

scaling relationships using BC test results. One example of such a scaling relationship was
presented by an interested party during the framework document public meeting and submitted
to the rulemaking docket (Pacific Gas and Electric Company (PG&E), No. 13, pp. 79), and is
illustrated in Figure 5.46 below. To scale the BC test results with battery energy using this
approach, DOE would vary the performance of the representative unit with battery energy in
proportion to a regression line, such as the one pictured in the figure.

Figure 5.46 24-hour efficiciency standard proposed by PG&E, which could be used to scale
the analysis of the representative unit to BCs with differing battery energies.
Because PG&E only provided suggestions on the variation of energy consumption with
battery energy, the other scaling relationships (energy consumption with regard to voltage, and
cost with regard to battery energy) would have to be developed through additional regressions, or
using one of the contrasting approaches discussed above.
5.6.8

Representative Product Class 1: Low Energy, Inductive

The low-energy, inductively charged product class includes BCs with a cradle that
couples inductively to the end-use product enclosure. These consist of a rectifier/low-pass filter,
the battery, and the end-use product (an electric toothbrush). This product class includes all
inductively-coupled BCs with a battery voltage less than 4 volts and a battery energy less than
100 watt-hours, though in actuality DOE has only found models with batteries with a voltage of
1.2 to 3.6 volts and energy of 0.5 to 1.8 watt-hours.

5-96

Despite the low shipments and low per-unit energy consumption in this product class,
DOE nonetheless decided to explicitly analyze it due to potential differences in design between
inductively-coupled chargers and other classes of BCs. Although DOE interviewed
manufacturers of inductively-coupled battery chargers, it was unable to perform product
teardowns due to a limited number of products in the market. This scarcity prevented DOE from
finding enough products with similar battery characteristics (energy and voltage) but with
differing efficiencies that would imply differences in design and warrant product teardowns.
5.6.8.1

Units Analyzed

The results of DOE tests of low-energy inductive BCs are pictured in Figure 5.47. Also
pictured are the results of tests conducted by Ecos Consulting and submitted to DOE by Pacific
Gas & Electric. As can be seen, there is substantial disparity between the baseline, intermediate,
and best-in-market units. In addition to the factor-of-two difference in battery energy between
the intermediate unit at approximately 0.9 watt-hours and the best-in-market at 1.8 watt-hours,
there is also a difference in voltage (not pictured) with batteries ranging from 1.2 to 3.6 volts.
Selecting a common point of comparison was not possible.

Figure 5.47 Test results for representative product class 1: Low Energy, Inductive.
DOE therefore turned to manufacturers of these products, who provided information on
units meeting the baseline through best-in-market CSLs for this product class. The estimated
performance characteristics of a representative unit based on an aggregate of the manufacturer

5-97

data is presented in Table 5.54, while the design options required to reach these levels of
performance are described in sections 5.6.8.2 through 5.6.8.4.
Table 5.54 Characteristics of typical units for product class 1 described by manufacturers
during interviews.
Rated
Battery
Energy

Rated
Battery
Voltage

Est.
Time in
Active

Est. 24-Hour
Charge
Energy

Est.
ActiveOnly
System

Est.
Maint.
Power

Est.
NoBattery
Power

UEC

Wh

Wh

Eff.

kWh/yr

CSL

Example
Application

Toothbrush

1.5

3.6

24.0

26.7

6%

1.2

0.5

10.0

Toothbrush

1.5

3.6

24.0

19.3

8%

0.8

0.4

7.0

Toothbrush

1.5

3.6

24.0

10.8

14%

0.4

0.2

3.5

Toothbrush

1.5

3.6

24.0

5.9

25%

0.2

0.1

1.5

5.6.8.2 CSL 0BaselineBaseline (UEC less than 10 kWh/yr)


Manufacturer Interview Results
The baseline unit represents BCs intended to inductively charge nickel-cadmium
batteries. Although many of these were designed as long as 10 years ago, they continue to be
sold in the market albeit at the lower price points.
Like other chargers in this product class, units meeting the baseline CSL were designed
for an electric toothbrush application. As such it is composed of two enclosures: a charging stand
that connects directly to AC mains and serves as the transmitter in the inductive link, followed
by a receiver and charge controller in the handle of the toothbrush. Because of the inductive
connectionwhich transfers power through magnetic coupling between transmitter and receiver
coils when they are in close proximityboth enclosures are sealed, allowing them to be used in
a wet environment.
What distinguishes the baseline units from units at the other CSLs is the slow charge rate
and lack of termination. The charge acceptance of a nickel cell decreases as the battery
approaches full charge, resulting in less charge in the cell for each unit of charge delivered by the
BC. Because it spends more time in this inefficient region of cell operation, the slow baseline
unit encounters significant active mode losses.
These active-mode losses are compounded by the lack of termination. Rather than turn
off the current to the battery after a full charge, the baseline unit continues to provide a small but
not insignificant current, resulting in an input power around 1 watt in active and maintenance
modes. No-battery power is less than 0.5 watts, while the resultant annual energy consumption is
less than 10 kilowatt-hours.

5-98

5.6.8.3 CSL 1Improved (UEC less than 7 kWh/yr)


Manufacturer Interview Results
Manufacturers have been able to improve the performance of the baseline charger by
substituting in energy-saving components and making other incremental improvements to the
design. Although many are still nickel-chemistry trickle chargers, units meeting CSL 1 save
energy over the baseline by using more efficient subcircuits and components, such as lower-loss
biasing of the input stage or higher-efficiency Schottky diodes in the inductive receiver. Schottky
diodes have half the forward voltage drop of conventional silicon pn-junction diodes, resulting in
half the losses.
The above improvement in biasing circuitry would reduce annual energy consumption
below 7 kilowatt hours and place these units into CSL 1. Specifically, the improvements in the
biasing circuitry would reduce the no-battery power to less than 0.5 watts. This, combined with
the component improvements, yields input power less than 0.75 watts in active and maintenance
modes, resulting in an annual energy consumption less than 7 kilowatt-hours.
5.6.8.4 CSL 2Best-in-Market (UEC less than 3.5 kWh/yr)
Manufacturer Interview Results
Currently, the best-in-market models in this product class ship with lithium-ion
toothbrushes. Even though lithium-ion cells have a nominal voltage of 3.6 volts (as opposed to
1.2 volts for the nickel cells described previously), the manufacturers considered the utility and
design of the lithium-ion best-in market unit to be comparable to the nickel units at the lower
CSLs.
Specifically, one can redesign the CSL 0 or CSL 1 units to meet CSL 2; however, it may
require improvements in all three modes of operation: active, maintenance, and no-battery.
Active-mode efficiency can be increased by improving the inductive coupling between the
transmitter and receiver by placing a resonant capacitor in series with the receiver coil or by
inserting additional ferrite material (which channels the magnetic field) in the charging stand.
Such improvements result in a 5 percentage-point higher active mode efficiency over a typical
CSL 1 unit..
However, according to manufacturers, the biggest gains can be made in the non-active
modes. Because it charges a lithium-ion battery, which cannot tolerate a continuous trickle
current, the best-in-market unit terminates following a full charge. Possible further
improvements include reducing the clock speed of the microcontroller or dimming the
informational LEDs located in the toothbrush handle. In total, these result in decreased input
power during maintenance mode below 0.5 watts, while leaving no-battery power unchanged.
The resulting annual energy consumption is less than 3.5 kilowatt-hours per year.

5-99

5.6.8.5 CSL 3Max-Tech (UEC less than 1.5 kWh/yr)


Manufacturer Interview Results
During interviews, manufacturers also discussed potential efficiency improvements
beyond the best-in-market level. According to manufacturers, technology options exist that can
decrease energy consumption in all three modes of operation.
First, manufacturers recommended improving the coupling between the toothbrush
handle and charger base through the addition of an extra coil, which could increase the
proportion of power transferred to the handle through resonance. Additionally, manufacturers
suggested slowing the microcontroller clock and dimming the status indicators (LEDs or LCD
backlights), which would decrease energy consumption in both active and maintenance modes.
Even with these improvements, the inductive connection would limit the overall system
efficiency in active mode to 25 percent. These design options, along with incremental
improvements in the charging base, can be expected to decrease the annual energy consumption
below 1.5 kilowatt-hours per year.
5.6.8.6 Estimate of Manufacturer Selling Price

Manufacturer Selling Price [$]

Because of an inability to perform meaningful teardowns, the MSPs for this product class
were based on the manufacturer interviews, and are pictured in Figure 5.48 and detailed in
Table 5.55.

$4.0

Mfr Interview Data

$3.5
CSL3

$3.0

CSL2

CSL1

CSL0

$2.5
$2.0

$1.5
$1.0

5
10
Energy Consumption [kWh/year]

15

Figure 5.48 Illustration of the MSP for this representative unit at each CSL.

5-100

Table 5.55 Comparison of MSPs for product class 1s representative unit obtained from
manufacturer interviews. No teardowns were performed for this representative unit.
MSP from
MSP from
UEC
Interviews Teardowns
CSL

kWh/yr

2009$

2009$

10.0

2.05

7.0

2.22

3.5

2.45

1.5

2.60

5.6.8.7 Scaling of CSLs within Product Class


DOE welcomes comment on potential strategies for scaling results of the representative
unit analysis to other BCs in this product class.
5.6.9

Representative Product Class 2: Low Energy, Low Voltage

The low-energy, low-voltage product class includes BCs for batteries below 4 volts and
below 100 watt-hours. It represents the low end of battery energy, and includes BCs for
telephony, personal care, and portable entertainment applications. DOE analyzed this
representative product class due to the large shipments of products in this category.
BCs in this representative product class charge batteries composed of lithium-ion, nickelcadmium, or nickel-metal hydride cells. Because the nominal voltage of a lithium-ion cell is
3.6 volts, there are no BCs for lithium-ion batteries in the lower portion of the voltage range;
instead in this range DOE found BCs for nickel-based batteries with one or two cells (nickel cells
have a nominal voltage of 1.2 volts). At or above 3.6 volts, either battery chemistry can be used;
however, applications that are smaller, lighter, or more expensive tend to use lithium-ion
batteries, while the remainder use nickel batteries.
To ensure an analysis applicable to BCs for lithium-ion as well as the two nickel
chemistries, DOE selected a 3 watt-hour, 3.6 volt charger as its representative unit. That energy
and voltage is characteristic of batteries of all three chemistries as well as a wide range of
applications.
5.6.9.1 Units Analyzed
DOE tested 19 units for this representative product class; their annual energy
consumption, assuming a shipment-weighted product class-average usage profile, is pictured in
Figure 5.49. For comparison, the figure also includes the results of tests conducted by Ecos
Consulting.2 Of these, DOE chose four units for further evaluation and teardowns based on their
efficiency test results and internal design. Detailed information regarding these four units is
presented in Table 5.56.

5-101

Annual Energy Consumption [kWh]

Increasing Efficiency

25

Cordless Phones
Cordless Phones/Answering Devices
Mobile Phones
Cordless Rechargeable Can Openers
Digital Cameras
Portable Video Game Systems
Ecos Tests
Rep. Unit
Efficiency Improvement Path

20

15

10

0
0

4
6
8
10
Measured Battery Energy [Wh]

12

14

Figure 5.49 Test results for representative product class 2: Low Energy, Low Voltage.
As can be seen in Figure 5.49, the majority of products tested in this product class have a
battery energy less than 5 watt-hours, with the exception of the cluster of Ecos test points at
approximately 10 watt-hours. These correspond to tests of universal battery chargers for
standard-sized AA or AAA batteries, performed with the highest number of batteries these units
could charge (typically 4). However, Ecos also tested these same units with the smallest
allowable number of batteries, whereupon the measured battery energy ranged from 0.9 to
2.2 watt-hourswithin the main cluster of battery energies. DOE therefore focused its analysis
on units at approximately 3 watt-hours to make its analysis applicable to the widest range of BCs
in this class, including the universal battery chargers.
From these, DOE selected several units for further analysis based on their measured
energy consumption, detailed in Table 5.56. These units were selected to span a wide range of
efficiencies available in the market. The only BCs that lie outside this efficiency distribution are
the Ecos test units with annual energy consumption above 10 kilowatt-hours per year.
In its test results database, Ecos lists these all as cordless phones/answering devices, and
despite a thorough search, DOE was unable to find any cordless phones with energy
consumption at those levels. This discrepancy between the lowest-efficiency units tested by DOE
versus Ecos may be due to the vintage of the units tested by Ecos (2007, prior to the efficiency
impacts of EISA 2007) and the potential energy consumption of answering machine features. In
either case, establishing the baseline CSL at the energy consumption of a current BC (with an
EISA-compliant EPS) unburdened by any additional non-battery charging functionality, is the
preferred approach.

5-102

Table 5.56 Detailed test results for the units corresponding to the CSLs for representative
product class 2: Low Energy, Low Voltage.
Battery
Energy
CSL
0
1

2
3

Application
Cordless
Phone
Digital
Camera
Digital
Camera
-

Wh
-

Rated
Battery
Voltage

Est.
Time in
Active

24-Hour
Charge
Energy

ActiveOnly
System

Wh
-

Eff.
-

Maint.
Power

NoBattery
Power

W
-

UEC
kWh/
yr
-

2.9

3.6

21.4

33.5

10%

1.4

0.3

6.0

2.8

3.7

2.6

8.8

45%

0.1

0.1

1.2

2.8
-

3.7
-

2.0
-

7.5
-

54%
-

0.1
-

0.1
-

1.0
-

Manufacturer interviews further influenced DOEs decision to set the baseline for this
representative unit at a higher UEC than that of the least efficient unit tested. Several
manufacturers described line-frequency BCs currently in production with annual energy
consumption higher than 5.1 kilowatt-hours. Typical performance of these manufacturer
described baseline units can be seen in Table 5.57, next to the performance of higher-efficiency
manufacturer units.
Table 5.57 Characteristics of typical units for product class 2 described by manufacturers
during interviews.

CSL
0

Example
Application

Rated
Battery
Energy

Rated
Battery
Voltage

Est.
Time in
Active

Est. 24-Hour
Charge
Energy

Est.
ActiveOnly
System

Est.
Maint.
Power

Est.
NoBattery
Power

UEC

Wh

Wh

Eff.

kWh/yr

3.6

22.0

41.1

8%

1.8

0.7

10.0

Power Tool
Cordless
Phone

3.6

14.5

31.3

14%

1.1

0.4

6.0

Cell Phone

3.6

6.9

9.8

79%

0.4

0.1

1.2

DOE seeks comment on whether 10 kilowatt-hours per year is a valid baseline for this
representative uniti.e., whether there are significant shipments of this representative unit at this
UECdespite the apparent lack revealed through testing. Finally, since there are units with this
UEC within this product class, but not with the same characteristics as the representative unit
(they have slightly different battery energies), DOE also seeks comment on whether teardowns
of units with slightly lower or higher measured battery energy could nonetheless be used in the
analysis of this representative unit.

5-103

5.6.9.2

CSL 0Baseline (UEC less than 10 kWh/yr)

Test and Teardown Results


As mentioned above DOE did not tear down any products at the baseline CSL, due to a
lack of product with this efficiency at the representative unit battery voltage and energy. For the
later analyses (life-cycle cost and national impacts analyses), DOE therefore used the results
from the test and teardown unit at CSL 1 as the baseline, with the result that there was no
significant impact from a standard set at CSL 1. See chapters 8 and 10 for more information.
Manufacturer Interview Results
Although manufacturers discussed the full range of efficiency for this product class,
chargers at the baseline efficiency in practice are limited to infrequently charged applications
used around the home, such as cordless telephones and handheld vacuum cleaners. The least
efficient chargers in this class are slow, which makes them useful primarily for applications that
are used infrequently and otherwise left for long periods in maintenance mode. Because they
have to dissipate more heat, the less-efficient baseline chargers are also typically more bulky
than chargers at higher CSLs, again relegating them to those same applications, which need not
be portable.
The baseline charger typically consists of a trickle charger for a nickel battery. The power
conversion is performed by a line-frequency transformer followed by a half-wave or bridge
rectifier. The impedance of the transformer windings typically performs charge control.
As in the case of the inductive baseline unit, the slow charge rate results in large
recombination losses in the battery. Active-mode efficiency (including the battery) can be as low
as 10 percent, though typically varies between 30 and 60 percent. Maintenance mode power, due
to the constant trickle current, is around 2 watts, while no-battery powerdriven primarily by
magnetization losses in the transformercan be as high as 1 watt. This performance results in an
annual energy consumption below 10 kilowatt-hours.
5.6.9.3 CSL 1Improved (UEC less than 6 kWh/yr)
Test and Teardown Results
Disassembly of the CSL 1 unit for this product class revealed a line-frequency design (the
baseline topology described by manufacturers), though with certain efficiency improvements,
likely prompted by the mandatory EPS efficiency standards put in place by EISA 2007. For
instance, the charger wall adapter uses higher-grade transformer steel, resulting in a particularly
low no-load power (0.29 watts), which results in the equally low no-battery power listed in
Table 5.56.
Furthermore, the rectifier is composed of four 1N5818 power Schottky diodes. With
roughly half the forward voltage drop, and consequently half the losses, of conventional silicon
diodes, these Schottkies contribute to the baseline units improved efficiency in active mode,
especially in relation to some of the units tested by Ecos Consulting in 2007. As no dedicated
charge control components were found, the transformer inductor windings are assumed to
perform this function.

5-104

As can be seen in Figure 5.50(b), the initial portion of the input power waveform during
charge mode displays the decaying exponential characteristic typical of line-frequency slow
chargers. This characteristic is due to the relationship between charger output current and battery
voltage. In this type of charger, the current is proportional to the battery voltage per Eq. 5.2.

VOUT VBATT
ROUT
,

Eq. 5.2
Where:
I is the charge current;
VOUT is the charger output voltage;
VBATT is the battery voltage; and
ROUT is the output impedance of the charger, typically due to the transformer windings,
which performs charge control.
As the battery recharges, its voltage increases, decreasing the charge current and slowing
the rate of charge, resulting in an exponential input power characteristic. Even though the input
power for this charger is fairly low, the long charge time, and in particular the time spent at high
state-of-charge, a shortcoming cited by manufacturers during interviews, contributes to the active
mode losses in this charger.

5-105

(a)

(b)

Figure 5.50 Input power waveform during charge and maintenance modes for the baseline
unit. The full 24-hour plot (a) shows some increased activity in maintenance, possibly associated
with phone operation, while the six-hour fragment (b) shows the exponential characteristic
typical of slow chargers.
Manufacturer Interview Results
In general, manufacturers agreed with the teardown results, and suggested that small
improvements to the baseline unit (CSL 0) can be made through the use of termination timers or
improved transformer steel. Nonetheless, they mentioned that a significant decrease in annual
energy consumption can only be achieved by moving to a switched-mode topology.
By using a high-frequency switched-moderather than line-frequencypower supply to
perform power conversion, a manufacturer not only decreases power conversion losses but can
also increase charge rate, j decreasing recombination losses in the battery. Recombination losses
occur inside the nickel batterywhich remains popular among many of the applications served
by BCs in this product classespecially as it approaches full charge. By spending less time at
this high state-of-charge, less of the charger output energy is consumed by recombination,
resulting in lower overall losses.
The typical active mode efficiency (including losses in the battery) of a switched-mode
BC meeting this CSL are above 50 percent, though some line frequency chargers still exist, with
active mode efficiency as low as 14 percent. Maintenance mode power ranges between 0.5 and
1.3 watts, while no-battery mode power is typically no more than 0.4 watts. The annual energy
consumption is less than 6 kilowatt-hours.
j

Because a switched-mode power supply is more efficient, it dissipates less heat, allowing it to be placed in a
smaller package. Alternatively, one could also keep the same package, but increase output power until the power
dissipated as heat reaches the limits of the original package. This higher output power allows one to charge a battery
at a much faster rate (e.g., 3 hours instead of 16) while maintaining the same charger size.

5-106

5.6.9.4 CSL 2Best-in-Market (UEC less than 1.2 kWh/yr)


Test and Teardown Results
To decrease the energy consumption of the baseline unit, a manufacturer needs to move
to a switched-mode power supply, which is one of the efficiency improvements in the digital
camera chargers at CSL 2, the performance of which is detailed in Table 5.56.
The two digital cameras differ both in physical form and internal construction. The first
charger listed in Table 5.56, which is slightly less efficient, charges an internal lithium-ion
battery and is therefore composed of two enclosures connected by a USB cablea wall adapter
and additional electronics inside the camera body. The more efficient unit charges an external
lithium-ion battery, and has both power supply and charge control circuitry integrated into a
single enclosure that plugs directly into mains. The best-in-market CSL can therefore be met by
chargers for both external as well as internal batteries, cable losses notwithstanding.
Furthermore, the CSL can also be met by chargers for other applicationsone of the other units
at CSL 2 torn down by DOE was a video game charger, though that particular unit has been
excluded from the analysis because a microcontroller that was used for both battery charging and
other functions inflated its cost past the point where it was comparable to the other chargers.
An interesting difference between the two camera chargers lies in their choice of power
supply topology. As can be seen in Figure 5.51, the switched-mode power supply for the
slightly-less efficient charger does not have an integrated controller IC, relying instead on a
ringing-choke converter (RCC) topology, which nonetheless manages to achieve an active mode
efficiency of 75 percent when tested according to the EPS test procedure. The power converter
circuit of the other charger features a flyback design; its efficiency independent of the battery is
unknown because of the inability of the EPS test procedure to test chargers that connect directly
to an external battery.

5-107

Figure 5.51. The top and bottom of the slightly less-efficient best-in-market camera charger.
Note the lack of any integrated circuit controller.
As can be expected, both chargers have additional circuitry beyond the power supply
because of the sensitive nature of lithium-ion batteries. In particular, lithium battery charging
requires a microcontroller or dedicated charge-management IC, as can be seen in Figure 5.52,
increasing the chargers cost. Nonetheless, these charge management ICs are necessary to
provide not only safety, but also low energy consumption during maintenance and no-battery
modes expected of best-in-market BCs.

5-108

Figure 5.52 PCBs of digital camera chargers meeting CSL 1. Note the highlighted
microcontroller in a black 14-pin surface-mount package in the left photograph. Similar
functionality is provided in the other charger by the dedicated charger IC highlighted at the right.

In particular, the charge and maintenance waveforms for the two camera chargers are
pictured in Figure 5.53 and Figure 5.54, below. As can be seen, the maintenance mode power is
approximately 0.1 watts for both chargers. This is likely accomplished through a combination of

5-109

minimal no-load/low-load losses in the wall adapters and effective power management in the
end-use application using the ICs mentioned previously.
Finally, the no-battery power for the best-in-market chargers is also approximately
0.1 watts. Since no-battery power is in many cases analogous to EPS no-load power,k the two
can be compared, and based on the responses obtained during the EPS manufacturer interviews
(summarized in section 5.4 specifically5.4.8), these power levels are near the theoretical
minimum achievable with todays technology.

Figure 5.53Fragment of the input power waveform during charge and maintenance modes
for the less efficient CSL2 digital camera charger.

In many cases, such as all integral-battery products powered by wall adapters, the two are equivalent.

5-110

Figure 5.54 Fragment of the input power waveform during charge and maintenance modes
for the more efficient CSL2 digital camera charger.
Manufacturer Interview Results
Manufacturers can achieve the best-in-market CSL through component improvements in
the power supply stage such as lower-loss diodes or switching transistors. Manufacturers also
recommended upgrading the integrated-circuit (IC) controller that manages the operation of the
power supply, permitting lower-power operation in low- and no-load modes through cycle
skipping and frequency foldback.
Even lower energy consumption can be achieved through additional circuit
improvements, though the gains at this stage are limited. The few manufacturers that addressed
annual energy consumption below 2.5 kilowatt-hours mentioned synchronous rectification as a
way to improve active-mode efficiency by a percentage point, though they characterized this as
more of a max-tech option. Further improvements in the IC controller could decrease the nobattery power to 0.1 watts, though still lower power levels could result in a power supply that is
not sufficiently responsive to the demands of the load (the downstream components of the
charger as well as any additional functionality).
Using the above strategies, manufacturers can increase the system-wide efficiency above
70 percent. BCs meeting this CSL also exhibit improvements in maintenance and no-battery
power, which are below 0.4 and 0.1 watts, respectively, consistent with that found through
teardowns. These energy consumption figures, combined with the average usage profile for this
product class lead to a typical annual energy consumption below 1.2 kilowatt-hours.
5.6.9.5 CSL 3Max-Tech
Test and Teardown Results
No analysis of the maximum-technologically feasible efficiency for this representative
product class has been performed to date.

5-111

Manufacturer Interview Results


As mentioned before, DOE testing demonstrated that some units currently in the market
offer comparable performance to the max-tech units described by manufacturers. Consequently,
the manufacturer max-tech design options for this and some other representative units have
been subsumed under the best-in market CSL 2, while DOE continues to investigate more
efficient designs.
5.6.9.6 Estimate of Manufacturer Selling Price

Manufacturer Selling Price [$]

The MSPs derived from teardowns and manufacturer interviews are pictured in Figure
5.55and detailed in Table 5.58 for each CSL. The manufacturer MSPs are an aggregate of the
costs provided by manufacturers, appropriately marked up, while the teardown MSPs correspond
to the MSP of the cordless phone charger for CSL 1 and the average of the two camera chargers
for CSL 2. No unit meeting the baseline CSL was torn down.

$15
Mfr Interview Data
Test and Teardown Data

CSL2

$12
$9
CSL1

CSL0

$6
$3
$0

5
10
Energy Consumption [kWh/year]

15

Figure 5.55 MSPs for each CSL for representative unit 2.


Table 5.58 Comparison of MSPs for product class 2s representative unit obtained from
manufacturer interviews versus those developed through teardowns.
MSP from
MSP from
UEC
Interviews Teardowns
CSL

kWh/yr

2009$

2009$

10.0

0.62

6.0

1.09

0.62

1.2

11.71

2.62

5-112

5.6.9.7 Scaling of CSLs within Product Class


As mentioned in section 5.6.9.1, above, this product class includes univeral battery BCs,
which typically charge either 1, 2, or 4, AA or AAA standard-sized batteries. Furthermore, DOE
received from Pacific Gas & Electric the results of tests performed on several of these chargers.
These test results, illustrated in Figure 5.56, could serve as the basis for scaling the analysis
results to all chargers within this representative product class.

Figure 5.56 Results of universal battery charger test results performed by Ecos Consulting
with batteries of varying energies. All test results for a given charger are performed with
batteries with the same voltage.
5.6.10 Representative Product Class 4: Low Energy, High Voltage
The low-energy, high-voltage product class includes BCs for batteries from 10 to 48 volts
and less than 100 watt-hours, in particular the majority of BCs used with notebook computers,
power tools, hedge and weed trimmers, and handheld vacuums. The notebook computer BCs
tend to cluster around 11 volts and 20 to 55 watt-hours, while the remaining applications tend to
span the voltage range between 10.8 and 24 volts, though with lower battery energies.
(Manufacturer views regarding this division are summarized in section 5.6.5.) DOE analyzed this
representative product class because the mid-range shipments numbers (millions to tens of
millions of units per year) combine with mid-range unit energy consumption to result in
significant shipment-weighted energy consumption.
The battery voltage and energy for all the applications in this product class, obtained
from the market survey, are illustrated in Figure 5.57.

5-113

Figure 5.57 Illustration of the limited overlap in BC characteristics between notebook


computers and motor-operated appliances in the low-energy high-voltage product class
(based on advertised battery voltage and energy).
To ensure an analysis representative of all the applications in this product class, DOE
chose as its representative unit a BC with a 10.8 volt, 20 watt-hour battery, which falls at the
overlap of the notebook computer and power tool ranges discussed above. The limited overlap
between the major applications in this product class made it difficult for DOE to test BCs from
both applications with the same battery voltage and energy. However, even though the
characteristics of the BCs tested during this analysis differed somewhat from the representative
unit and each other, these variations are small compared to the large differences in efficiency
seen in the next section.
5.6.10.1

Units Analyzed

DOE tested eight units for this representative product class; their annual energy
consumption, assuming an average usage profile, is pictured in Figure 5.58. For comparison, the
figure also includes the results of tests conducted by Ecos Consulting. Of these, DOE chose four
units for further evaluation and teardowns based on their efficiency test results and internal
design. Detailed information regarding the measured performance of these four units is presented
in Table 5.59, while information on typical units provided by manufacturers appears in Table
5.60.

5-114

Annual Energy Consumption [kWh]

Increasing Efficiency

60
Power Tools
Notebook Computers
Ecos Tests
Rep. Unit
Efficiency Improvement Path

50
40

30
20
10
0
0

20
40
60
Measured Battery Energy [Wh]

80

Figure 5.58 Test results for representative product class 4: Low Energy, High Voltage
Table 5.59 Detailed test results for the units corresponding to the CSLs for representative
product class 4: Low Energy, High Voltage.
Battery
Energy

Rated
Battery
Voltage

Est.
Time in
Active

24-Hour
Charge
Energy

Active
Only
System

Maint.
Power

NoBattery
Power

Wh

Eff.

UEC
kWh/y
r

CSL

Application

Wh

Power Tool

16.3

12.0

0.6

167.5

56%

5.9

2.2

39.4

Power Tool

13.4

12.0

0.9

52.6

69%

1.4

1.4

10.5

Notebook
Computer

22.9

11.1

2.7

44.6

70%

0.6

0.1

6.1

Power Tool

15.2

12.0

0.6

29.3

84%

0.5

0.5

3.8

5-115

Table 5.60 Characteristics of typical units for product class 4 described by manufacturers
during interviews.
Rated
Battery
Energy

Rated
Battery
Voltage

Est.
Time in
Active

Est. 24Hour
Charge
Energy

ActiveOnly
System

Est.
Maint.
Power

Est.
NoBattery
Power

UEC

Wh

Wh

Eff.

kWh/yr

CSL

Example
Application

Power Tool

20

10.8

13.0

142.9

19%

3.4

1.2

39.4

Notebook
Computer

20

10.8

1.5

57.4

58%

1.0

0.5

10.5

Notebook
Computer

20

10.8

1.0

42.5

76%

0.7

0.2

6.1

As can be seen in Figure 5.58, all the BCs tested at the representative-unit battery energy
and voltage were used to establish the CSLs for the low-energy high-voltage product class. Two
BCs of comparable energy consumption were used as the basis for CSL 2 to make this analysis
applicable to both major applications in this representative product class.
Three BCs in Table 5.59 are standalone chargers for DIY power tools with external
batteries. They consist of a single enclosure that contains the power conversion and charging
circuitry and has a cup that holds the battery during charging. One or two charge indicator
LEDs are also present. The chargers are powered through a non-detachable AC line cord.
The fourth BC, intended for notebook computer applications resides in two enclosures.
One is a wall adapter with detachable AC line cord, while the other is the notebook computer
itself, containing an integral battery. The wall adapter performs the AC/DC conversion,
outputting a regulated 20 volts DC, while the charge control and battery monitoring is performed
by circuits located on the computer motherboard, as shown in Figure 5.59.

Figure 5.59 Photographs of the top and bottom of the motherboard of the notebook
computer, with power management and battery charging components highlighted.

5-116

Despite their slightly different constructions, the four BCs provide comparable utility.
Each has an output power between 36 and 47 watts, and while the AC/DC converters of the
power tool chargers must convert to a lower voltage than that for the notebook computer
(approximately 12 volts versus 20 volts), the latter must perform two-stage voltage conversion
(from 120 volts AC to 20 volts DC to 12.3 volts DC). Lower output voltages and successive
conversions decrease the conversion efficiency, which impacts the energy consumed in active
mode. Because DOEs analysis focused on these real-world BCs, it is applicable to a wide-range
of BCs in this product class, regardless of their physical construction or inefficiencies due to the
requirements of the end-use application.
5.6.10.2

CSL 0Baseline (UEC less than 39.4 kWh/yr)

Test and Teardown Results


DOE chose a 12 volt BC with a measured battery energy of 16.3 watt-hours and annual
energy consumption of 39.4 kilowatt-hours as its baseline unit for this product class. The energy
consumption of this unit was the highest of all the 12 volt units that DOE tested, and furthermore
exceeded the energy consumption of the majority of BCs in this product class, regardless of
battery voltage and energy.
As can be seen in Figure 5.58, however, Ecos Consulting tested a power tool charger
with a measured battery energy of 15.7 watt-hours that had an annual energy consumption of
58.7 kilowatt-hours. Furthermore, because this was an 18 volt charger, one could expect its
energy consumption to be even higher if the same design were applied to the lower-voltage
representative-unit battery. Nonetheless, DOE did not use this or a similar BC as its baseline due
to its slow charge rate: unlike most BCs in this product class, this charger did not terminate after
a few hours of charging, and at the end of the 24 hour test was still drawing over 10 watts
(compared to approximately 12 watts initially). Such behavior is atypical of BCs in this product
class.
During the market survey, DOE compiled information on 57 cordless power tools
currently popular in the market. While not all the manufacturers listed the charge time, of the 28
that did, the maximum charge time was 3 hours. Given this information, DOE does not expect
slow chargers for power tools to be available in the future as consumers seek shorter charge
times. Therefore, DOE considered the annual energy consumption of this slow 18 volt,
15.7 watt-hour charger to be an outlier and set its baseline at the energy consumption of the least
efficient fast charger, described below.
The baseline unit charges nickel-cadmium batteries for power tool use. Its design consists
of a flyback converter, with some additional circuitry to control the charging of the battery. For
example, in contrast to AC/DC converters that serve as voltage sources (e.g., EPSs), the baseline
BC unit has two optocouplers. As in an EPS, the first is likely intended to send feedback signals
related to the output voltage from the secondary to the primary sides of the power converter,
while the second likely controls the charging current.
As can be seen in Table 5.59, the baseline unit has a low 24-hour energy efficiency
(approximately 10 percent), calculated by dividing the battery discharge energy by the energy
consumed by the charger over a 24-hour charge and maintenance cycle. However, this low

5-117

performance is not due to low power conversion efficiency in the AC/DC converter. In fact, if
one examines only the initial portion of the 24-hour test, when the battery is actively charging,
up to the point where the charger enters maintenance mode, the charger energy consumption is
only 29 watt-hours, for an active -only battery charging system efficiency of:
ACTIVE

EBATT 16.3 watt- hours


E ACTIVE 29 watt- hours

56.2 percent

Eq. 5.3
where:
ACTIVE

is the active-only efficiency;

EBATT is the energy recovered from the battery during discharge; and
E ACTIVE is the energy consumed by the charger during active mode, as defined through

the examination of the input power measurements presented in Figure 5.58.


The inefficiency, therefore, occurs in maintenance mode. Because the charge time for this
BC is only 0.6 hours, the high input power in maintenance mode (almost 6 watts) dominates the
24-hour combined charge and maintenance efficiency. This high energy consumption is not due
to overhead losses in the charger, but rather to an inefficient maintenance strategy, which uses
high-current pulses to periodically top off the battery. While topping off the battery may be
beneficial over the long run, it should not be implemented immediately after the battery has
finished charging.
As can be seen in Figure 5.60, the input power of the baseline unit following the full
charge is periodic in time, with each of the spikes in the figure lasting approximately 2 minutes,
consuming an additional 1.1 watt-hours, and delivering a significant portion of that energy to the
battery after it has already been fully charged. Not only does this result in inefficient operation,
but the resulting elevated temperature also decreases battery lifetime.

5-118

Figure 5.60 The input power of the baseline unit during charging and maintenance modes.
The overall efficiency of the baseline unit is further compromised by the choice of
components. None are optimized for energy efficiency. For example, the design relies on
standard pn-junction diodes for output rectification, while the switching is performed by a
generic UC3842-series PWM controller, which draws up to 17 milliamperes with a supply
voltage above 10 volts DC and a frequency in the 50 kilohertz range. In sum, the baseline unit
has an annual energy consumption of 39.4 kilowatt-hours.
Manufacturer Interview Results
Despite the decreasing popularity of slow chargers for power tool operation,
manufacturers focused on them when discussing the baseline CSL. Nonetheless, not all
applications can tolerate the long charge times of slow chargers: given an ideal battery, they
should take 10 hours to charge; however, due to recombination losses in the nickel battery,
typical charge times are on the order of 16 hours.
On the contrary, BCs for professional power tools typically recharge the battery in
1 hour, while those for notebook computers in 2 to 3 hours. Manufacturers described such fast
chargers as relying on a high-frequency switched-mode topology because the size of a linefrequency transformer that could provide the power necessary to recharge the battery in such a
short time would be too large to be economical or user-portable. Nonetheless, not all switchedmode power supplies are naturally efficient. Some manufacturers have been producing the same
BCs for upwards of 10 years, and neither the power supply nor the charge control circuitry has
changed to take advantage of the latest energy efficient technology.
According to manufacturers, these inefficient fast BCs typically consist of a non-isolated
switched mode power supply (all other components of the charger, including the battery charging
terminals, are then inaccessible to the user to prevent electrical shock). Iron-core isolated power
supplies are similarly inefficient. Finally, the remainder of the charger design has not been
optimized with regards to energy efficiency. This results in either high overhead losses or
preventable losses from other system components (particularly in applications where the charge
is embedded inside the end-use product).

5-119

As a result, the active mode efficiency of these battery chargers (including the efficiency
of the typical nickel-cadmium batteries) is around 55 percent. The maintenance mode power
ranges between 2 to 6 watts (with the higher number comparable to the teardown unit) while the
no-load power ranges between 1 to 2 watts. Annual energy consumption is less than
39.4 kilowatt-hours.
5.6.10.3

CSL 1Improved (UEC less than 10.5 kWh/yr)

Test and Teardown Results


The BC unit associated with CSL 1 is also a fast power tool charger, with a 12 volt,
13.4 watt-hour battery. In contrast to the baseline unit, the CSL 1 unit does not continue to
charge the battery in maintenance. The lithium-ion battery would not tolerate it.l Instead, as can
be seen in Figure 5.61, the input power to the unit drops to 1.4 watts. As can be seen in Table
5.59, this is also the input power to the unit in no-battery mode, when the battery is removed
from the charger. Since the BC draws the same power whether or not the battery is present, this
indicates that maintenance mode power is already as low as possible. Further improvements can
only be made by focusing on the overhead power dissipation of the electronics, such as the on
board microcontroller, op-amp, voltage reference, etc., which impact both no-battery and
maintenance modes.

Figure 5.61Fragment of the input power waveform during charge and maintenance modes
for the CSL 1 unit.
The conversion efficiency of this BC is also improved over the baseline unit. Unlike the
baseline, the CSL 1 unit uses a Schottky diode for rectification, which has a lower voltage drop,
resulting in lower power dissipation during charging and higher active mode efficiency.
Nonetheless, the design evaluated for CSL 1 continues to use UC3842-series PWM controller to
drive the switching FET. As mentioned previously, the high supply current requirements of this

Unlike nickel-based batteries, lithium-ion batteries cannot tolerate overcharge so lithium-ion chargers must
terminate following full charge.

5-120

controller limit the maximum active mode efficiency and minimum maintenance and no-battery
power.
Manufacturer Interview Results
Manufacturers proposed numerous design options for improving the performance of the
baseline charger. These ranged from moving to a switched-mode topology and termination for
the slow chargers, to various incremental circuit improvements for the fast chargers.
To improve the efficiency of the slow charger, manufacturers recommended
implementing terminationcutting off the charge current following the rechargeusing either a
timer or state-of-charge sensor. Although that would decrease maintenance mode current, that
alone may not be enough. To decrease energy consumption in all three modes, manufacturers
also recommended improvements to the line-frequency transformer core steel (resulting in lower
magnetization losses) or a complete transition from line-frequency to high-frequency switchedmode topology, similar to that currently used for fast chargers.
For fast chargers, which already used switched-mode designs, but often inefficiently,
manufacturers recommended incremental circuit improvements. According to interviewees,
replacing any silicon output diodes with Schottkies could significantly decrease energy
consumption. This is consistent with what was seen in the teardowns. Other incremental
improvements included improvements to the passives, such as inductors with lower parasitic
resistances and lower-leakage filter capacitors.
As an alternative path to reaching CSL 1, some manufacturers recommended more
efficient charge control, for example using buck converters as opposed to linear regulators to
adapt the power supply output to the requirements of the battery charging stage.
Finally, some of the manufacturers focused on minimizing maintenance and no-battery
power, which manufacturers mentioned are often indistinguishable from each other, a fact also
evidenced by DOE testing. For instance, one manufacturer suggested adding an electro
mechanical switch that would disconnect the charger once the battery is removed (only practical
for chargers with external batteries)), while another spoke of replacing the components that
enable communication between the BC and the battery-pack electronics. The newest generation
of these components, although more expensive, has lower standby losses, leading to lower BC
energy consumption in standby and maintenance modes.
The above design options, either singly or in combination, could improve the energy
efficiency of the baseline unit to the CSL 1. In particular, the active mode efficiency of the entire
battery charging system could increase from 55 percent to 65 or 70 percent, while the
maintenance power could decrease below 2 watts (consistent with testing), and the no-battery
power below 1 watt. Implementing the automatic disconnect switch could eliminate no-battery
power to 0 watts. The resultant annual energy consumption is less than 10.5 kilowatt-hours.

5-121

5.6.10.4

CSL 2Best-in-Market (UEC less than 6.1 kWh/yr)

Test and Teardown Results


As mentioned previously, DOE identified two units that exhibited an annual energy
consumption superior to all other BCs tested in this product class. These two units demonstrate
that this level of energy consumption is commercially achievable for both power tools and
notebook computers, the two major applications in this product class. Because of their lithiumion batteries, both BCs must use efficient maintenance-mode strategies that cut off current to the
battery, such thatas seen previouslytheir input power in maintenance is equal to the input
power in no-battery mode and due entirely to the overhead power of the control electronics.
In the case of the power tool charger, these electronics are fairly efficient and include a
Schottky diode for rectification, reducing energy consumption during power conversion
associated with charging. The BC also features a more efficient PWM controller, which
integrates the controller and FET into a single package with a maximum supply current of 1.3 to
2.0 milliamperes, depending on the switching frequency. This is an order of magnitude less than
that of the controller in the baseline and CSL 1 units.
The design of the BC for notebook computer applications demonstrates alternate methods
of reaching CSL 2 and maintaining energy consumption below 4.2 kilowatt-hours per year.
Unlike the power tool chargers examined earlier, this BC features a two-stage architecture, with
an external power supply with an industry-standard regulated 20 volt DC output, followed by a
battery charger embedded inside the application.
The EPS uses Schottky diodes for rectification, resulting in lower forward voltage drops
across the rectifier and higher conversion efficiency in active mode. The EPS also features a
more efficient PWM controller for driving the switching FET, with a maximum supply current of
1.2 to 2.0 milliamperes, depending on the switching frequency.
The charger portion of the BC internal to the notebook computer uses a dedicated
integrated circuit (IC) for charge control. This IC consists of a DC/DC buck converter, which
reduces the 20 volt input from the EPS to a lower voltage suitable for the 11.1 volt lithium-ion
battery, while monitoring the current and voltage to the battery to ensure safe charging.
Although a two-stage architecture typically introduces further losses into a battery
charging system,m these are lowered by using a switched-mode second stage. Synchronous
rectification used on the output of this second stage makes it even more efficient and puts this
BC on par with the best-in-market single-stage BC. Nonetheless, a two-stage solution does offer
some benefits. As can be seen in Table 5.59, the input power during no-battery mode is six times
less than that during maintenance mode.
This is not due to any detection of the presence of the battery (another design option), but
is simply achieved by disconnecting the entire second stage together with the battery. Because
the notebook computer has an integral battery, a user does not remove it from the application for
m

The efficiencies of each stage are multiplied together to arrive at the system-efficiency, such that two 90-percent
efficient stages will result in a system with only 81 percent efficiency.

5-122

recharging; rather, the user connects the entire applicationthe computer, the battery, and the
second-stage battery charging electronicsto the EPS. In no-battery mode, the inverse is true,
with the notebook computer disconnected and the EPS the only part of the product connected to
the AC line. Therefore, the second stage and other potential sources of loss from the computer
power-management circuitry are no longer present, decreasing the input power compared to that
in maintenance mode. Such a strategy could have broader applicability beyond notebook
computers, as additional electronics continue to be packaged with detachable batteries for such
applications as power tools, and was in fact called out by manufacturers during interviews as a
possible efficiency design option.
Manufacturer Interview Results
Manufacturers agreed that no slow-charger design could meet the best-in-market CSL. As
a result, all applications that currently use slow chargers would need to convert to fast chargers
to meet this CSL; however, this should only improve consumer utility. Furthermore, the majority
of chargers at this CSL utilize lithium-ion batteries, which manufacturers agreed offer increased
electro-chemical efficiency during charge and discharge. The drawback of lithium-ion is that
because of their sensitive nature, these batteries require constant supervision using dedicated
safety circuits, a burden which continues to make nickel-metal hydride batteries competitive in
terms of both efficiency and cost.
Nonetheless, some manufacturers stressed that a move to fast-charging, switched-mode
power supplies and strategies typically used with lithium-ion batteries (e.g., limited maintenancemode current) would not be enough to meet CSL 2. While some claimed that this level of energy
consumption was not achievable given todays technology, others presented a range of advanced
methods necessary to decrease energy consumption to equal the most efficient units currently in
the market.
These strategies include adaptive charge control that tailors the charge to the exact
condition of the battery, resulting in minimal overcharge. Additionally, the power supply could
be further improved with the addition of green-mode switching controller ICs, capable of
reducing losses at low load, and synchronous rectification, which would reduce losses at high
load. When used in concert, the two technologies would have the effect of flattening out the
efficiency curve (which typically sags at low and high loads, with a peak in the middle), such
that the BC would remain efficient over a wider range of its typical operating range.
Nonetheless, the problem of energy consumption in the non-active modes (maintenance
and no-battery) persists. Although green-mode controllers can reduce the energy passing through
the transformer to the secondary side of the power supply and the charge-control stage, they
cannot eliminate the primary-side losses due to any remaining transformer magnetization losses
in the transformer as well as the quiescent current to the controller itself.
To address this issue, manufacturers proposed adding a second, 1 watt power supply
whose sole purpose would be monitoring the state of the output and input and powering the main
power supply, as necessary. Because of its small output power, this secondary power supply
would have minimal overhead losses, resulting in low maintenance and no-battery power for the
charger as a whole. Short of including an entire power supply (something that has been
implemented in practice in the case of multiple-voltage non-Class A EPSs), manufacturers also

5-123

proposed a monitoring circuit on the primary side that could respond to a signal from across the
primary/secondary isolation barrier and enable the controller IC to provide current to the load.
Again, such a circuit would result in best-in-market performance in non-active modes. Finally, as
for CSL 1, manufacturers also mentioned the possibility of adding an electro-mechanical switch
to detect the presence of the battery, causing disconnection of the primary in no-battery mode.
Once the circuits have been redesigned per the above strategies, manufacturers mentioned
that improving the parasitic resistances of line-filtering inductors and capacitors could result in
some additional reductions in energy consumption, though compliance with emissions
requirements could be a challenge.
These changes could be expected to result in system-wide active-only efficiency of
76 percent, maintenance power below 0.75 watts, and no-battery power below 0.25 watts. The
annual energy consumption would be below 6.1 kilowatt-hours.
5.6.10.5

CSL 3Max-Tech

Test and Teardown Results


No analysis of the maximum-technologically feasible efficiency for this representative
product class has been performed to date.
Manufacturer Interview Results
Again, manufacturers did not present any designs that surpassed those currently found in
the market. Consequently, many of the manufacturer max-tech design options have been
subsumed under the best-in market CSL, and DOE continues to search for truly max-tech
designs.
5.6.10.6

Estimate of Bill-of-Materials Costs

iSuppli, a DOE contractor, tore down the four BCs corresponding to CSL 0CSL 3 for
this product class. iSuppli estimated the manufacturing costs (i.e., the materials, assembly, and
test costs) of the components related to battery charging. For example, in the case of the
notebook computer at CSL 2, iSuppli only evaluated the circuitry related to battery charging and
supplying power to the battery charger, including battery charging components mounted to the
computer motherboard.
Figure 5.62 shows both manufacturer and teardown cost-efficiency curves for this
product class. To construct the teardown curve, DOE took the cost and efficiency of the two
units at CSL 0 and CSL 1 directly. To accommodate any remaining differences in cost between
power tool and notebook computer BCs, DOE averaged the cost of the two most efficient units
to arrive at a CSL 2 cost representative of both applications. The efficiency point for CSL 2 was
that of the less efficient unit, ensuring that both units would be able to meet this CSL. These data
are also presented in Table 5.61 Comparison of MSPs for product class 4s representative
unit obtained from manufacturer interviews versus those developed through teardowns.

5-124

Manufacturer Selling Price [$]

$15
Mfr Interview Data
Test and Teardown Data

$12
$9

CSL0

$6
CSL2

$3
CSL1

$0

10
20
30
40
Energy Consumption [kWh/year]

50

Figure 5.62 The efficiency related manufacturing cost of the four commercially analyzed
BCs in this product class.
Table 5.61 Comparison of MSPs for product class 4s representative unit obtained from
manufacturer interviews versus those developed through teardowns.
MSP from
MSP from
UEC
Interviews Teardowns
CSL

kWh/yr

2009$

2009$

39.4

3.79

3.79

10.5

9.52

6.76

6.1

12.68

7.44

5.6.10.7

Scaling of CSLs within Product Class

Because both notebook computer and power tool manufacturers often sell accessory
batteries of varying capacities, one possible way to determine scaling relationships for this
product class would be to retest the units described previously in section 5.6.11.1 with larger and
smaller replacement batteries. Similarly, some power tool BCs can charge batteries of varying
voltages (for compatibility with multiple product lines); again tests with several batteries of
varying voltage may lead to the development of an appropriate scaling relationship.
5.6.11 Representative Product Class 5: Med. Energy, Low Voltage
The medium-energy, low-voltage product class includes BCs for batteries less than
24 volts and 100-3000 watt-hours. Batteries that meet these criteria typically have a sealed lead

5-125

acid chemistry and are used for medium-sized motor-operated products such as lawn mowers,
marine trolling motors, and wheelchairs. Because of the higher capacities of these batteries,
chargers in this product class typically have much higher output powers than chargers for the
majority of consumer products (higher energy transferred to the battery over a similar period of
time). As a result, they employ different power converter designs (forward and half-bridge as
opposed to flyback) than those lower-power chargers.
Unlike the lower-energy high-volume consumer product BCs represented by the earlier
product classes, these BCs tend to use standard-sized 6 or 12 volt lead-acid batteries, typically
purchased separately from the BC and the end-use application. This presents two problems for
the purposes of the analysis.
First, because of standardization of voltages by application, batteries for wheelchairs tend
to have a particular voltage and energy, while those for marine motors, another. Therefore, in
contrast to earlier product classes, where the choice of battery was at the discretion of the
manufacturer and there was a wide range of both battery voltage and energy for each application,
it was difficult to find an overlap in terms of battery voltage and energy between the different
applications. DOE therefore focused its representative-unit analysis solely on one application,
namely the marine charger, though with the understanding that the analysis could be applied
more broadly.
Second, because the batteries are purchased separately from the BC, tests of a BC may
produce different results depending on which battery is chosen. DOE intends to address this
issue through its parallel BC test procedure rulemaking and welcomes suggestions from
stakeholders.
5.6.11.1

Units Analyzed

DOE tested four chargers for 12 volt, sealed lead acid batteries typically used with
scooters, wheelchairs, and marine trawling motors. Because the batteries for these applications
vary only in terms of capacity (measured in ampere-hours), the same charger can hypothetically
be used for any of these applications.n Therefore, where possible, DOE tested chargers in this
product class with batteries for both wheelchair and marine applications. The results of these
tests, as well as ones performed by Ecos Consulting, are summarized in Figure 5.63, below.

In practice, however, the smallest (and also least expensive) charger will be selected which can recharge a given
battery within a required time. For example, since wheelchair batteries have a capacity around 15 ampere-hours (at
the 5-hour rate) while marine batteries have a capacity around 60 ampere-hours, chargers for the former have a two
to four times smaller output power and size than the latter.

5-126

300
Wheelchairs
Marine Battery Charger

Annual Energy Consumption [kWh]

Increasing Efficiency

250

Ecos Tests
Rep. Unit
200

Efficiency Improvement Path

150

100

50

0
0

200

400

600

800

1000

1200

Measured Battery Energy [Wh]

Figure 5.63 Test results for representative product class 5: Medium Energy, Low Voltage.
As can be seen in the figure, the two marine chargers that DOE selected as its teardown
units for analysis lie in the middle of marine charger efficiency, with shipment-weighted average
energy consumption in the range of approximately 100 to 150 kilowatt-hours per year. On the
other hand, Ecos tests with a roughly 800 watt-hour sealed lead-acid battery resulted in a range
of energy consumption from 65 to over 400 kilowatt-hours per yeara much wider range.
At the high end of energy consumption, DOE purchased and examined a charger with
similar characteristics as the inefficient one tested by Ecos, but determined that this charger was
not suitable for testing with the sealed lead-acid batteries used for marine application, because of
its lack of voltage control, which could lead to overcharge and damage to the battery. (The BC
was likely intended for use with automotive starting batteries). Therefore, DOE discarded the
results of this test as representing a suitable baseline for this product class.
At the low end, DOE again purchased and examined chargers with similar characteristics
as those tested by Ecos, but was unable to replicate equivalent measurements. These
discrepancies may have been due (1) shortened tests, (2) inefficiencies in the battery used for the
DOE tests, and/or (3) an insufficiently broad variety of chargers.
Although the recently-proposed DOE active-mode battery charger test procedure
specifies a 24-hour measurement period for the charge and maintenance mode tests, several of

5-127

the DOE tests were shortened for the sake of expediency when the battery charger was observed
to enter a constant lower-power mode characteristic of maintenance. Nonetheless, it is possible
that the behavior of the charger would not have stayed constant had the charger continued
operating past this early termination, perhaps entering an even lower-power state, resulting in
lower measured maintenance mode power.
Secondly, an inefficient battery may have caused the lower-than-expected active mode
efficiencies of the battery charging system. Because DOE used a different sealed lead-acid
battery than Ecos, it is conceivable that its electrochemical charge and discharge efficiencies
were lower, resulting in lower results for all the chargers tested. Again, how to deal with battery
variation will be addressed in the test procedure.
Finally, it is possible that none of the four marine chargers that DOE tested performed at
the highest levels available in the market. Regardless of the cause of the higher-than-expected
energy consumption of the chargers tested by DOE, additional testing and teardowns should
resolve this issue. In the meantime, DOE has used the units it has torn down described in
detail in Table 5.62to only represent the baseline and improved CSLs (CSL 0 and CSL 1).
The remaining CSLs have been supplied by manufacturers through interviews. The typical
performance data provided by manufacturers is listed in Table 5.63.
Table 5.62 Detailed test results for the units corresponding to the CSLs for representative
product class 5: Medium Energy, Low Voltage.
Battery
Energy

Rated
Battery
Voltage

Est.
Time in
Active

24-Hour
Charge
Energy

Active
Only
System

Maint.
Power

NoBattery
Power

UEC

kWh

Eff.

kWh/yr

CSL

Application

Wh

Marine

830.7

12.0

12.8

2.04

46%

21.2

20.1

202.7

Marine

762.1

12.0

15.9

1.65

49%

11.9

11.6

159.6

5-128

Table 5.63 Characteristics of typical units for product class 5 described by manufacturers
during interviews.
Rated
Battery
Energy

Rated
Battery
Voltage

Est.
Time in
Active

Est. 24Hour
Charge
Energy

Est.
ActiveOnly
System

Est.
Maint.
Power

Est. NoBattery
Power

UEC

Wh

kWh

Eff.

kWh/yr

CSL

Example
Application

Marine

800

12

13.0

1.83

33%

10.0

10.0

202.4

Marine

800

12

12.0

1.62

39%

10.0

10.0

158.4

Marine

800

12

8.0

1.30

55%

0.1

0.1

100.0

Marine

800

12

8.0

1.18

78%

0.0

0.0

75.0

5.6.11.2 CSL 0Baseline (UEC less than 202.4 kWh/yr)


Test and Teardown Results
The baseline marine charger tested by DOE is typical of low-cost chargers in this class,
and is composed of a line-frequency transformer with a center-tapped secondary winding and
two rectifier diodes. In addition, there is a small charge management circuit board with a BTW69
silicon-controlled rectifier (SCR) in series with the DC output lead, pictured in Figure 5.64. This
subcircuit is responsible for the linear input power characteristic evident in Figure 5.65. Finally,
the unit contains rather small diodes in button packages, which are held against the heatsink with
a plastic clamp. Over time, the plastic is likely to change shape decreasing the contact pressure
between the diodes and the heatsink. While this can positively impact the efficiency of the unit
(for a given current, the diode forward voltage decreases as the temperature rises), in the long
run, the diodes will overheat leading to failure and a shorter lifetime.

5-129

(a)

(b)

Figure 5.64 Photographs of the baseline marine charger, showing (a) the line-frequency
transformer and (b) heatsink-mounted SCR. Note the difference in size and design
compared to the low-energy BCs.

5-130

Figure 5.65 Fragment of the input power waveform during charge and maintenance modes
for the baseline unit.
Manufacturer Interview Results
Manufacturers did not provide information regarding the baseline efficiency. Rather,
when developing the aggregate manufacturer response, Navigant Consulting extrapolated the
relationship between cost and efficiency at higher CSLs down to the baseline efficiency. The
extrapolation is explained in section 5.6.5.
5.6.11.3

CSL 1Improved (UEC less than 158.4 kWh/yr)

Test and Teardown Results


The improved (CSL 1) marine charger analyzed by DOE also features a large linefrequency transformer with a center-tapped secondary winding and two rectifier diodes. The
large PCB, containing a microcontroller and other through-hole parts provides charge control,
and can be adjusted using a front-panel switch depending on the type of battery (flooded or
sealed lead-acid).
On the whole, this charger appears much more durable than the baseline unit, with a
heavy conformal coating on the board (increasing the longevity of the unit in damp and dirty
environments) and a heavier-than-usual aluminum bracket, which serves as a heatsink for the
diodes.

5-131

(a)

(b)

Figure 5.66 Photographs of the improved marine charger, showing (a) the line-frequency
transformer and (b) complex charge-control PCB.
Despite providing the additional functionality of charging two battery types, this unit
performs better than the baseline, charging the battery in less time with fewer losses. The input
power characteristic of the improved unit in active and maintenance modes is presented in
Figure 5.67.

5-132

250

Input Power [W]

200
150
100
50
0
0:00:00

6:00:00

12:00:00

Test Time [h]

Figure 5.67Fragment of the input power waveform during charge and maintenance modes
for the CSL 1 unit.
Manufacturer Interview Results
Again, manufacturers did not provide information regarding the improved efficiency,
resulting in a manufacturer-derived relationship that is a continuation of that found at the higher
CSLs.
5.6.11.4

CSL 2Best-in-Market (UEC less than 100 kWh/yr)

Test and Teardown Results


DOE did not tear down any more efficient medium-energy, low voltage chargers beyond
the two described in Section 5.6.11.2 and 5.6.11.3, above. Nonetheless, DOE would expect that
charge management could reduce power consumption at end of charge. Since chargers in this
product class spend large amounts of time in maintenance mode, such improvements could result
in significant annual energy savings.
Manufacturer Interview Results
To meet the best-in-market CSL, manufacturers suggested using a silicon-controlled
rectifier (SCR) charger with a best-in-class line-frequency transformer, which can be up to 50%
more expensive than that used at CSL 1. In addition to transformer improvements, manufacturers
suggested using the SCRs to perform both rectification and control by placing them directly in
the center-tapped half-wave rectifier; or if using a full-wave bridge rectifier, replacing the silicon
pn-junction diodes with Schottkies.
Alternatively, manufacturers also suggested foregoing the SCR topology entirely, in
favor of a high-frequency switched-mode design, which has the added benefit of lighter weight
due to smaller transformer size. Finally, to completely eliminate energy consumption in

5-133

maintenance and no-battery modes, manufacturers suggested including a relay to disconnect the
primary side of the battery charger from mains. A typical manufacturer unit at CSL 2 had a
system-wide active-only efficiency of more than 55 percent, maintenance power of 0.1 watts,
and no-battery power of 0.1 watts, for an annual energy consumption lower than 100 kilowatthours.
5.6.11.5

CSL 3Max-Tech (UEC less than 75 kWh/yr)

Test and Teardown Results


No analysis of the maximum-technologically feasible efficiency for this representative
product class has been performed to date.
Manufacturer Interview Results
To achieve the maximum technologically feasible efficiency, resulting in an annual
energy consumption of less than 10 kilowatt-hours, manufacturers did not propose any novel
topologies. Instead they provided incremental improvements to the SCR and switched-mode
topologies developed for CSL 2, above.
Since, as described above in section above, non-active energy consumption in
maintenance and no-battery modes could be eliminated entirely for some application through the
use of a relay, manufacturers focused on further reducing the energy consumption in active mode
to meet CSL 3. These could be achieved by increasing the efficiency of the transformer through
further investment in core steel (reducing magnetization losses) and winding copper (reducing
resistive conduction losses). Similar reductions in resistive losses could be made throughout the
rest of the charger by the widths of the conductive traces on the PCB or the gauge of the
connecting cables.
These improvements could result in a system-wide active-only efficiency approaching
80 percent (above 90 percent, excluding the battery), maintenance and no-battery power at
0 watts, and annual energy consumption below 10 kilowatt-hours.
5.6.11.6

Estimate of Manufacturer Selling Price

As before, iSuppli, a DOE contractor, tore down the two BCs at CSL0 and CSL 1 for this
product class, and estimated the cost of the electronic components. These were subsequently
marked up to reflect manufacturers costs, resulting in the partial teardown curve in Figure 5.68.
The figure also shows the aggregated manufacturer MSPs. Although manufacturers provided
costs only at CSL 2 and CSL 3, the cost-efficiency relationship they developed was extrapolated
to the lower CSLs, where it corresponded will with the teardown results. The detailed results for
both manufacturers and teardowns are presented in Table 5.64.

5-134

Manufacturer Selling Price [$]

$150
$135
$120
$105
$90
$75
$60
$45
$30
$15

Mfr Interview Data


Test and Teardown Data

CSL3
CSL1

CSL0

CSL2

50

100
150
200
Energy Consumption [kWh/year]

250

Figure 5.68 Cost-efficiency results at each CSL for representative unit 5.


Table 5.64 Comparison of MSPs for product class 5s representative unit obtained from
manufacturer interviews versus those developed through teardowns.
MSP from
MSP from
UEC
Interviews
Teardowns
CSL

kWh/yr

2009$

2009$

202.7

18.48

18.48

159.6

27.46

21.71

100.0

64.14

75.0

127.00

5.6.11.7

Scaling of CSLs within Product Class

Because chargers in this product class are sold independently from the battery, they can
be tested with a variety of batteries of different capacities. By retesting the units discussed above
using different batteries, as well as testing additional units, appropriate scaling relationships can
be developed.
5.6.12 Representative Product Class 7: High Energy
The high-energy representative product class includes BCs for batteries with an energy
greater than 3000 watt-hours. Whereas the BCs included in product class 6 were intended to
charge batteries for a variety of applications, the high-energy batteries associated with chargers
in product class 7 are only used for golf cars and utility vehicles. Furthermore, these high-energy
batteries use a flooded or wet lead-acid construction, meaning that the batteries are free to vent to
the outside air in case of excessive gas buildup. Although this requires additional care in

5-135

handling (e.g., the batteries should not be tipped or left exposed to the elements), it does make
the batteries more resilient to overcharge and results in looser tolerances (and lower costs per
watt of output power) on the chargers. These differences led DOE to place these chargers in a
separate product class, though some manufacturers claimed there was broad similarity between
all chargers for medium and high energy batteries. DOE requests input on its decision to
separate these two product classes despite manufacturer comments suggesting that it may not be
necessary.
5.6.12.1

Units Analyzed

Because of the high energy of golf car batteries compared to those for other consumer
products, DOE had difficulty performing the necessary tests and teardowns for inclusion in the
preliminary analysis. Therefore, DOE relied on manufacturer interviews to create a costefficiency relationship for this product class. Performance figures typical of those presented by
manufacturers for units meeting each CSL are shown in Table 5.65.
Table 5.65 Characteristics of typical units for product class 7 described by manufacturers
during interviews.
Rated
Battery
Energy

Rated
Battery
Voltage

Est.
Time in
Active

Est. 24Hour
Charge
Energy

Est.
Maint.
Power

Est.
NoBattery
Power

UEC

kWh

Est.Active
Only
System
Eff.

Wh

kWh/yr

CSL

Example
Application

Golf Car

3750

48

8.0

5.90

64%

3.6

3.7

290.0

Golf Car

3750

48

8.0

5.61

68%

3.3

3.2

250.0

Golf Car

3750

48

8.0

5.24

72%

2.9

2.6

200.0

3
Golf Car
3750
48
8.0
4.86
78%
2.6
2.0
150.0
* Note that the battery energy is measured over a 5-hour discharge, in accordance with the proposed BC test
procedure.

5.6.12.2

CSL 0Baseline (UEC less than 290 kWh/yr)

Manufacturer Interview Results


Baseline golf car chargers use ferro-resonant (FR) or silicon-controlled rectifier (SCR)
topologies, though FR is falling out of favor because of recent increases in the cost of the highgrade steel due to competition with the global electricity transmission and distribution industry.
According to manufactures, the cost of an FR transformer can be up to 80 percent of the cost of
the charger, such that these swings in materials prices directly impact the manufacturer selling
price of the charger, causing a shift toward SCR.
In either case, the chargers use standard silicon pn-junction diodes for rectification,
resulting in charger-only (i.e., excluding any losses in the battery) active mode efficiencies of
approximately 80 percent. Manufacturers differed regarding the prevalence of energy
improvements intended to reduce energy consumption in the non-active modes. Although some
of the chargers described had no-battery and maintenance power greater than 5 watts, others used
relays that cut off power to the primary after the charge has finished, interrupting the flow to the
5-136

battery and resulting in an input power of 0 watts in maintenance and no-battery modes. The
annual energy consumption for a baseline BC in this product class is less than 290 kilowatthours.
5.6.12.3

CSL 1Improved (UEC less than 250 kWh/yr)

Manufacturer Interview Results


To reduce the energy consumption of the chargers in active mode with the goal of
reaching CSL 1, manufacturers suggested either incremental improvements or a wholesale
redesign of the BC to take advantage of a switched-mode topology.
Potential incremental improvements involved the substitution of Schottky diodes, which
have half the forward voltage drop of conventional diodes, in the bridge rectifier. Although
further incremental improvements could be made, the more cost-effective alternative would be a
redesign of the charger for one with a high-frequency switched mode power supply. The benefits
of such a change include not only higher efficiency, but also lower weight, higher power factor,
and usability across a wider range of input voltages.
While the Schottkies are likely to increase the active-mode efficiency by 1 percentage
point, the move to a switched-mode topology provides a significant increase in efficiency, by
around 5 percentage points, resulting in charger-only efficiencies around 85 percent. Input power
in maintenance and no-battery modes still depends on the manufacturers willingness to install a
relay. Regardless of which design options are implemented, the annual energy consumption is
less than 250 kilowatt-hours.
5.6.12.4

CSL 2Best-in-Market (UEC less than 200 kWh/yr)

Manufacturer Interview Results


Meeting the best-in-market CSL would require a transition to a switched-mode topology,
with charger-only active-mode efficiency above 85 percent and input power in maintenance and
no-battery modes below 3 watts. As mentioned above, switched-mode topologies may have some
additional utility advantages over the baseline and improved SCR chargers, and as long as the
maintenance and no-battery power is reduced, they may meet CSL 2 with annual energy
consumption below 200 kilowatt-hours.
5.6.12.5

CSL 3Max-Tech (UEC less than 150 kWh/yr)

Finally, manufacturers also speculated on ways of further reducing the energy


consumption of current best-in-market units, though, according to manufacturers, no further
improvements in topology exist, with additional incremental improvements offering diminishing
returns.
Manufacturers first proposed increasing the widths of all conductors in the battery
charger as a way of reducing resistive losses, similarly to the methods in product class 5,
described in section 5.6.11.5. These conductors include the PCB traces, the cables connecting the
charger to the battery, and the transformer windings. However, these improvements have
limitations: space inside the charger is limited and wires with a thicker gauge will be bulkier and

5-137

less flexible. Furthermore, wider wires in the transformer will push the windings away from the
core. The resultant spacing will cause more magnetic flux to leak away from the transformer and
reduce the coupling from the primary to the secondary winding, counteracting any improvement
in efficiency due to lower resistance. At best, such improvements could result in a charger-only
active-mode efficiency slightly above 90 percent.
Further improvements would include upgrades to the magnetic components (including
the transformer cores), replacing diodes with FETs (synchronous rectification), and finally
decreasing the losses of the FETs, in order of increasing costs. Such techniques could increase
the active-mode efficiency to around 95 percent in active mode (excluding the battery), though at
very high cost. These changes, however, would be necessary to decrease the annual energy
consumption below 150 kilowatt-hours.
5.6.12.6

Estimate of Manufacturer Selling Price

As mentioned above, DOE did not perform teardowns of units in this product class; therefore,

Manufacturer Selling Price [$]

Figure 5.69 shows only the manufacturer cost-efficiency curve shown at each CSL. Detailed
cost information is also presented in Table 5.66.
$250
$225
$200
$175
$150
$125
$100
$75
$50
$25
$0

Mfr Interview Data

CSL3
CSL2
CSL1
CSL0

100

150
200
250
300
Energy Consumption [kWh/year]

350

Figure 5.69 Manufacturer cost efficiency relationship for representative unit 7.

5-138

Table 5.66 Comparison of MSPs for product class 7s representative unit obtained from
manufacturer interviews versus those developed through teardowns.
MSP from
MSP from
UEC
Interviews Teardowns
CSL

kWh/yr

2009$

2009$

290

79.01

250

94.94

200

127.05

150

192.32

5.6.12.7

Scaling of CSLs within Product Class

Although the propulsion systems of golf cars (the dominant application in this product
class) can differ between manufacturers, requiring different battery voltages and capacities, these
differences (36 versus 48 volts and 200 versus 150 ampere-hours, at the 20-hour rateo) should not
substantially impact the achievable efficiencies or cost. Basically, the power levels in this
product class are so high that the fixed losses are unlikely to have a differential impact on the
performance of a BC charging a 36 volt, 200 ampere-hour battery versus a 48 volt, 150 amperehour battery.
DOE therefore does not expect to scale the CSLs throughout this product class, and is
considering using a flat standard level applicable to all units, equal to the CSLs presented in
Table 5.66.
5.6.13 Representative Product Class 10: Low Energy, AC Output
The final product class analyzed includes BCs that are a part of uninterruptible power
supplies (UPSs). UPSs are battery-operated products that provide backup power to other
electronic products in case of a power outage. As such, they differ from other BC products in
that they are never used except in case of emergency and have additional circuitry downstream
of the battery (an inverter) to provide AC output to the electronic appliances protected.
Navigant Consulting specifically interviewed manufacturers of uninterruptible power
supplies to determine the impact of the inverter and other specialized circuitry on the costefficiency relationship of BCs for these applications. The results of these interviews, as well as
product tests and teardowns, are summarized in the sections below.
5.6.13.1

Units Analyzed

DOE purchased five UPSs from an online retailer for testing and possible teardowns.
Like transformers and other power-electronic components, UPSs are rated in terms of maximum
o

Lead-acid battery capacity is typically advertised at the 20-hour rate. Because of current-dependent losses and
dynamic behavior in the battery, the actual capacity will be lower when measured at a 5-hour rate (consistent with
the proposed BC test procedure), lower still at the 23 hour rate (the typical rate of discharge during use).

5-139

volt-ampere (VA) output, which is a measure of total power (both real and reactive) they can
provide, and is equal to the maximum power drawn by a load divided by its minimum power
factor. Four of the five units had output between 500 and 650 volt-amperes, which manufacturers
later confirmed was a popular range within their consumer/small business product lines. Two of
the unitsone at 250 volt-amperes and another at 550 volt-ampereswere not included in the
further analysis because they provided additional functionality, described below, that inflated the
apparent battery charger energy consumption and confounded the comparison of cost and
efficiency. All these units used 12-volt valve-regulated lead-acid (VRLA) batteries with a rated
capacity of 7 to 9 ampere-hours (at the 20-hour rate). This battery configuration is, again, typical
of UPSs used by consumers.
These units were tested according to the proposed DOE test procedure. As was the case
for the other product classes, the per-mode measurements obtained using the test procedure were
then weighted according to the usage profile to obtain an annual energy consumption. In the case
of UPSs, which spend the vast majority of their time in maintenance mode, this simply involved
multiplying the maintenance power by number of hours per year (8760).
The results of testing and weighting appear in Figure 5.70, which compares DOE results
with those conducted by Ecos Consulting. Unlike in many of the other product classes analyzed,
the battery characteristics should not significantly impact the energy consumption of UPS BCs,
as maintenance mode powerwhich is largely independent of the battery characteristicsis the
only metric that contributes to annual energy consumption.

Annual Energy Consumption [kWh]

Increasing Efficiency

250

200

150

100
Uninteruptible Power Supplies
Ecos Tests
Rep. Unit
Efficiency Improvement Path

50

0
0

50

100

150

200

Measured Battery Energy [Wh]

Figure 5.70 Test results for the low-energy, AC-output representative product class.

5-140

Instead, any differences in energy consumption appear to be the result of charger design
or additional functionality. Note, for instance, the least efficient UPS tested by DOE, with an
annual energy consumption of approximately 180 kilowatt-hours. This corresponds to over
20 watts in maintenance mode, but is not due to any gross inefficiencies in the charger, but rather
to line voltage regulation, which conditions the power delivered to the load by passing it through
the battery.
The same applies to the DOE-tested unit with an annual energy consumption of
approximately 65 kilowatt-hours, which also adjusts to variations in the input voltage. The highannual energy consumption Ecos units at the top of the plot are assumed to provide similar
additional functionality.
As such, it is difficult to separate the energy consumption required for battery charging
and maintenance from that due to device operation. The situation is analogous to trying to
measure the energy consumption of a BC for a notebook computer with the notebook computer
turned on and operational. Furthermore, unlike in a notebook computer, the charging and power
conditioning functions may not be separablei.e., turning off the unit may also disconnect the
power provided to the battery. These units were therefore excluded from further analysisuntil
DOE finalizes how it intends to address products with additional functionality in its proposed
battery charger test procedure.
Of the remaining UPSs tested, two were selected for further analysis. The test results for
these are listed in Table 5.67, while specific design options for these UPSs are described in the
sections that follow. As before, Navigant Consulting, a DOE contractor, also interviewed
manufacturers of UPSs to gain further insight on the cost-efficiency relationship. The interviews
proved particularly informative because of the integration of the BC into the end-use product in
this product class; the manufacturers provided further insight into the impact of the inverter and
other specialized circuitry on the cost-efficiency relationship of these BCs. The typical
performance of units at each CSL as reported by manufactures is presented in Table 5.68, while
further discussion is presented in the sections below.
Table 5.67 Detailed test results for the units corresponding to the CSLs for product class
10: Low-Energy AC Output.
Battery
Energy

Rated
Battery
Voltage

Est.
Time in
Active

24-Hour
Charge
Energy

ActiveOnly
System

Maint.
Power

NoBattery
Power

UEC

Wh

Wh

Eff.

kWh/yr

CSL

Application

UPS

78.0

12.0

24.0

149.9

42%

2.2

19.6

UPS

51.7

12.0

17.3

93.0

44%

0.7

6.4

5-141

Table 5.68 Characteristics of typical units for product class 10 described by manufacturers
during interviews.
Rated
Battery
Energy

Rated
Battery
Voltage

Est.
Time in
Active

Est. 24Hour
Charge
Energy

Est.
ActiveOnly
System

Est.
Maint.
Power

Est.
NoBattery
Power

UEC

Wh

Wh

Eff.

kWh/yr

CSL

Example
Application

UPS

70

12

24.0

24%

2.2

19.6

UPS

70

12

24.0

50%

0.7

6.4

UPS

70

12

24.0

62%

0.5

4.0

UPS

70

12

24.0

84%

0.2

1.5

5.6.13.2 CSL 0Baseline (UEC less than 19.6 kWh/yr)


Test and Teardown Results
DOE disqualified two UPS from further analysis as their relatively high energy
consumption was likely due to additional functionality (power conditioning) unrelated to battery
charging and it selected the next best-performing UPS as its baseline unit for analysis.
The input power waveform for this charger during active and maintenance modes is
pictured in Figure 5.71, and shows the constant-current, constant-voltage characteristic typical
of lead-acid battery chargers.p The resultant efficiency is approximately 42 percent, including
losses in the battery, and it is made possible by a more efficient switched-mode power supply.
The resultant annual energy consumption of this unit is 19.6 kilowatt-hours.

During the first (flat) part of the curve, the charger provides a fixed current. The input power nonetheless increases
because battery voltage is increasing as the battery charges, and the power to the battery (to the input power). The
second portion sees the charger maintaining a constant voltage across the battery, with exponentially decreasing
charge current, and, consequently, input power.

5-142

Figure 5.71 Input power waveform during charge and maintenance modes for the CSL 1
unit.

5-143

Figure 5.72 Photograph of the PCB of the baseline unit for this product class. The BC
circuitry is mainly to the left.
Manufacturer Interview Results
The baseline CSL corresponds to the lowest efficiency UPS tested by DOE, except for
those that performed power conditioning and therefore consumed power in excess of that
required for battery charging and any mainstream UPS functions. According to manufacturers,
this level can be met by a variety of charger designs, including a fairly inefficient line-frequency
charger with charger-only active-mode efficiency less than 50 percent. Since for the majority of
U.S. consumers power outages are rare, BCs for UPSs seldom operate in active mode. Therefore,
the active mode efficiency does not impact the BCs annual energy consumption. Furthermore,
since the battery is internal and the AC line cord attaches directly to the UPS (such that no
component of the BC can remain connected to the line), the no-battery mode is not applicable to
this product class. Therefore, the only metric influencing annual energy consumption is input
power during maintenance mode.

5-144

According to the interviews, typical input power during maintenance mode for linefrequency BCs is approximately 2 watts, due to typical linear converter standby losses such as
magnetization losses in the transformer or leakage in the bulk capacitor. The parasitic losses in
the transformer windings, which are used to control charge current to the battery, also play a
role.
Additional losses specific to the UPS, and which cannot be disabled, include the power
draw of a microcontroller that monitors not only the battery charging, but also the state of the AC
line and communications and control lines connected to a personal computer (the typical
electronic product connected to such a UPS). The result is an annual energy consumption of less
than 19.6 kilowatt-hours.
5.6.13.3

CSL 1Improved (UEC less than 6.4 kWh/yr)

Test and Teardown Results


The improved teardown unit shows an even shorter charge time and lower active mode
power than the baseline, though those performance characteristics vary with battery size. (See
Figure 5.73) What does not vary, and ultimately determines the UPSs annual energy
consumption is maintenance mode power, which is very low (0.7 watts according to Table 5.67).
This low power is enabled by a switched-mode power supply, which not only has synchronous
rectification (paired transistors mounted to heatsinks on the secondary side of the power supply),
but also a low-power controller IC, with one third the quiescent current of the IC used in the
CSL 0 unit. The annual energy consumption of this BC is 6.4 kilowatt hours.

Figure 5.73 Input power waveform during charge and maintenance modes for the CSL 1
unit.
Manufacturer Interview Results
More efficient UPSs use a switched-mode topology as the input stage to the charger. The
switched-mode chargers have a similar cost as the line-frequency chargers meeting the baseline
CSL, but higher active-mode efficiency, typically higher than 85 percent.

5-145

However, as before, it is the maintenance mode power that impacts the energy
consumption of BCs for this application. The switched-mode charger, because it operates at a
much higher frequency, can use a smaller transformer, with consequently lower magnetization
losses. Overhead losses due to line monitoring and microcontroller operation can be further
reduced, resulting in a maintenance mode power less than 0.75 watt, resulting in annual energy
consumption less than 6.4 kilowatt-hours.
5.6.13.4

CSL 2 Best-in-Market (UEC less than 4 kWh/yr)

Test and Teardown Results


No analysis of the best-in-market efficiency for this representative product class has been
performed to date.
Manufacturer Interview Results
Although DOE was unable to identify any UPSs with annual energy consumption below
6.4 kilowatt-hours (the CSL 1 level), manufacturers indicated that incremental improvements can
decrease the energy consumption of the BC. In particular, energy-efficient power controller ICs
and improved FETs can be used to increase active-mode conversion efficiency of the BC to
greater than 80 percent (for a system-wide active-only efficiency above 65 percent).
Improvements to the microcontroller (typically used to control the charging process as well as
other aspects of the operation of the UPS) can also further decrease energy consumption in
maintenance below 0.5 watts. Together, these changes can reduce the annual energy
consumption below 4 kilowatt-hours.
5.6.13.5

CSL 3Max-Tech (UEC less than 1.5 kWh/yr)

Test and Teardown Results


No analysis of the maximum-technologically feasible efficiency for this representative
product class has been performed to date.
Manufacturer Interview Results
Finally, manufacturers speculated on further methods of decreasing the energy
consumption of these BCs beyond what is currently available in the market. Although higherefficiency switching converters for active-mode efficiency are available, they will not be
effectual in further reducing the annual energy consumption of the BC.
As in the other product classes, high-efficiency switched-mode controller ICs that
decrease energy consumption at low-load are a potential means of reducing maintenance power.
However, manufacturers also speculated on ways to completely disable the primary side of the
switched-mode power converter that is the first stage of the BC. By running from the integral
battery, the on-board microcontroller could completely shut down the input stage, enabling it
only when the battery had discharged past a desirable level or some external condition (such as
input voltage variation or communication with a personal computer) warrants.
According to manufacturers, such a strategy could result in maintenance mode power
below 0.17 watts, resulting in a max-tech UEC of 1.5 kilowatt-hours.

5-146

5.6.13.6

Estimate of Manufacturer Selling Price

Manufacturer Selling Price [$]

The results of iSuppli teardowns as well as aggregated manufacturer interviews, appear in


Figure 5.74. As mentioned above in section 5.6.13.2, teardown analysis revealed that the
baseline unit, despite its active mode inefficiency and high maintenance mode energy
consumption commanded a higher price than that at CSL 1. Again, DOE is concerned that this
units high energy consumption is due to some additional functionality that DOE did not take
into account, and seeks comment on the cost and efficiency of the baseline unit in this class.
Manufacturer selling price was calculated across all CSLs, even though manufacturers only
provided data for CSL 2 and CSL 3. Both sets of data are detailed in Table 5.69, below.

$5.0

Mfr Interview Data


Test and Teardown Data

$4.5
$4.0
CSL3

$3.5
CSL2

CSL0

$3.0
CSL1

$2.5
$2.0

5
10
15
20
Energy Consumption [kWh/year]

25

Figure 5.74 Teardown and interview MSPs at each CSL for representative unit 10.
Table 5.69 Comparison of MSPs for product class 10s representative unit obtained from
manufacturer interviews versus those developed through teardowns.
MSP from
MSP from
UEC
Interviews Teardowns
CSL

kWh/yr

2009$

2009$

19.6

2.76

2.76

6.4

3.93

2.11

4.0

4.25

1.5

4.64

5-147

5.6.13.7

Scaling of CSLs within Product Class

The market assessment and interviews with manufacturers revealed that uninterruptable
power supplies are the only applications covered by this product class. As discussed, the vast
majority of UPSs run entirely in maintenance mode and power consumption in this mode is
largely independent of the voltage and energy characteristics of the batteries. Additionally, DOE
believes that those units with excessive energy consumption as seen in DOEs (and ECOSs) test
results were because of added functionality provided by those UPSs. Therefore, DOE does not
intend to perform any scaling within this product class.
5.6.14 Extending the Analysis to Product Classes not Analyzed Directly
DOE focused its engineering analysis on the representative product classesproduct
classes with the highest number of shipments or the highest estimated energy consumption,
according to initial estimates based on the market survey. Nonetheless, there are still significant
shipments, and consequently, potential for energy savings, in the non-representative product
classes. The sections below describe how DOE extended the engineering analysis to these
product classes to evaluate the impact of potential standards.
For EPSs, DOE was able to directly model the cost-efficiency relationship for scaled
product classes using the curves developed for the representative product class. For BCs,
however, the cost-efficiency curves for each representative unit are functions of UEC and take
into account the usage profiles (described further in chapter 7) for the applications in that
representative product class. Therefore, DOE could not simply scale the BC cost-efficiency
results presented in section 5.6.8 through 5.6.13; rather, DOE had to scale and then re-weight
(according to the new usage profile) the per-mode data obtained through testing and
manufacturer interviews.
In section 5.6.7, above, DOE discussed some potential strategies for scaling the results of
the representative-unit analysis to units not analyzed directly, using engineering relationships or
data available in the market. Pending comment from interested parties on the applicability of the
listed methods, DOE used a simpler approach for the analysis of the scaled product classes. In
short, DOE assumed that the efficiency of the BCs tested or described by manufacturers would
not changerather, any changes in annual energy consumption would be due to the difference in
battery energy and usage between product classesq.
Thus, for example, when scaling the performance of a notebook computer BC for use
with scaled product class 3, DOE maintained the same charge time, active mode efficiency,
maintenance mode, no-battery mode, and off mode power, but now applied those to charging a
battery typical of BCs in that product class (e.g., those used with a portable DVD player). Since
the maintenance mode, no-battery mode, and off mode power were constant, only the energy lost
per charge would change.

For clarity, when it is stated that DOE performed an analysis it should be noted that Navigant Consulting handled
all sensitive information gleaned from manufacturer interviews.

5-148

Subsequently, these per-mode energy consumption metrics were weighted by the usage
profile of the scaled product class. Therefore, even though there may have been no great changes
in the per-mode energy consumption, the annual energy consumption may have changed
significantly due to differences in usage.
The detailed methodology and results for each representative product class are presented
in the sections that follow.
5.6.14.1

Scaled Product Class 3: Low Energy, Medium Voltage

This product class spans the gap between the low energy, low voltage and low energy,
high voltage product classes (i.e., BC product classes 2 and 4), which each have higher
shipments. Nonetheless, the applications in this product class are fairly similar to those in the
other two product classes: power tools, handheld vacuum cleaners, two-way radios, and portable
DVD players. The BC design is also similar to that of the other two product classes, consisting of
multi-cell nickel chargers (both fast and slow) and fast lithium-ion chargers.
The lithium-ion batteries used with chargers in this product class are multi-cell, like those
in product class 4. However, evaluations of several DVD players revealed none of the
complexities common to the design of the higher-voltage notebook computer batteries; rather,
they were more similar to the single-cell lithium-ion batteries for cellular phones and portable
media players. The most efficient BCs in the market for portable DVD players can also achieve
energy consumption as low as some of the low energy, low voltage units. Despite these
similarities in BC design, the usage in product class 3 was much lighter than in either of the
representative product classes, as can be seen in Table 5.68.
Table 5.70 BC Shipments-Weighted Average Usage Profiles for Product Classes 2, 3, and 4
BC Product
Active +
No Battery
Unplugged
Off
Charges
Class
Maintenance
(Standby)
Hours per Day
2
3
4

9.7
5.6
16.8

5.0
0.2
0.3

9.4
18.1
6.9

0
0.1
0.1

Number per
Day
0.58
0.22
0.88

Note: Usage times may not sum to exactly 24 hours due to rounding.

Because the products in this class are an amalgam of those in product classes 2 and 4,
DOE used the manufacturer cost-efficiency data for those representative product classes to
develop the cost-efficiency relationship for product class 3. As described above, Navigant first
scaled the energy loss per charge to account for the difference in battery energy between product
classes 2 and 4, and 3; then, Navigant Consulting applied the usage profile for product class 3 to
calculate the annual energy consumption of the manufacturer designs for product classes 2 and 4
when used with applications in product class 3. Nonetheless, DOE seeks comment on alternative
methods of determining costs and efficiencies for the low energy, medium voltage product class.

5-149

The baseline CSL for this product class was based on the lowest-performing results
provided by manufacturers, scaled to reflect the battery and usage of product class 3. CSLs 1
and 2 were based on the results of tests of BCs in product class 3 performed by DOE and Ecos
Consulting, again weighted to reflect typical use for product class 3. Since DOE did not tear
down any products in the scaled product classes, all costs are interpolations of the scaled
manufacturer cost-efficiency data. The results are summarized in Table 5.71 and Table 5.72 .
Table 5.71 Characteristics of typical units scaled for product class 3.
Rated
Battery
Energy

Rated
Battery
Voltage

Est.
Time in
Active

Est. 24Hour
Charge
Energy

Est.
ActiveOnly
System

Est.
Maint.
Power

Est.
NoBattery
Power

UEC

Wh

Wh

Eff.

kWh/yr

CSL

Example
Application

Power Tool

10

7.2

16.1

91.3

19%

4.7

1.0

10.0

Cell Phone

10

7.2

10.3

65.7

32%

2.5

0.6

5.4

Cell Phone

10

7.2

1.1

21.4

60%

0.2

0.2

1.0

Table 5.72 Resulting cost and efficiency relationship for BC product class 3.
MSP from
MSP from
UEC
Interviews Teardowns
CSL

kWh/yr

2009$

2009$

10.0

$3.16

5.4

$4.96

1.0

$12.99

5.6.14.2

Scaled Product Class 6: Medium Energy, High Voltage

This product class is similar to product class 5: medium energy, low voltage. BCs in this
product also charge sealed lead-acid batteries, though for a different set of applications. Because
of the general similarities between chargers at the higher battery energies, DOE used the costefficiency data obtained from manufacturers for product class 5, scaled according to the typical
battery energy and voltage of product class 6, and weighted by the usage profiles of product class
6 applications, such as lawnmowers and electric bicycles.
DOE based the baseline and improved CSLs (CSL 0 and 1) for this scaled product class
on the results of tests conducted on BCs in this class. In particular, DOE tested four lawnmower
BCs with a 408 watt-hour, 24 volt sealed lead-acid battery, and set the baseline to the UEC of the
worst-performing unit. Two of the other units reached an annual energy consumption of
approximately 40 kilowatt-hours, so DOE set the improved CSL at that level. The remaining two

5-150

CSLs (CSL 2 and 3) were based on the performance of manufacturer designs from representative
product class 5, scaled to the battery and usage of product class 6. Finally, the MSP was
calculated at each CSL by using the best-fit curve through the manufacturer data. The results are
summarized in Table 5.73 and Table 5.74.
Table 5.73 Characteristics of typical units scaled for product class 6.
Rated
Battery
Energy

Rated
Battery
Voltage

Est.
Time in
Active

Est. 24Hour
Charge
Energy

Est.
ActiveOnly
System

Est.
Maint.
Power

Est.
NoBattery
Power

UEC

Wh

Wh

Eff.

kWh/yr

CSL

Example
Application

Marine

384

36

20.0

720.8

55%

5.0

1.2

69.3

Marine

384

36

16.0

620.0

62%

0.5

0.5

41.9

Marine

384

36

16.0

524.6

74%

0.3

0.3

25.0

Marine

384

36

8.0

491.3

78%

0.0

0.0

18.0

Table 5.74 Resulting cost and efficiency relationship for BC product class 6.
MSP from
MSP from
UEC
Interviews Teardowns
CSL

kWh/yr

2009$

2009$

69.3

$18.48

41.9

$35.87

25.0

$76.83

18.0

$139.95

5.6.14.3

Scaled Product Class 8: Low Energy, 5 V DC Input

This product class contains small consumer electronic products typically recharged using
the 5 volt output of a computers USB port, in particular portable media players and personal
data assistants (PDAs). Interviews with manufacturers revealed that the BCs in this product class
are similar to those in representative product class 2, which are typically used with cellular
telephones, except that they do not require an EPS.
Therefore, to develop a cost-efficiency relationship for this product class, Navigant
Consulting subtracted out the typical EPS losses and costs at each CSL. To model these
quantities, Navigant Consulting used the EPS results for the 2.5 watt EPS, originally presented in
Table 5.29 and reproduced in part in Table 5.75 below.

5-151

Table 5.75 Cost and Efficiency Relationship for 2.5W EPS (Manufacturer Interviews)
CSL 0
CSL 1
CSL 2
CSL 3
Mfr Unit Efficiency [%]:
58.3%
67.9%
73.5%
74.0%
Mfr Unit No Load Power [W]:
0.500
0.300
0.100
0.053
MSP Difference with CSL0 [$]: $0.00
$0.05
$0.26
$0.32
To model the change in BC energy consumption without the presence of an EPS, such as
the one above, Navigant Consulting divided the active-mode efficiency provided by the
manufacturers at each CSL by the EPS efficiency. Similarly, Navigant Consulting subtracted
from the maintenance mode power and the no-battery mode power the EPS no-load power. Next,
Navigant Consulting subtracted out the incremental cost difference from the BC MSP so as to
eliminate the cost impact of improving the EPS on the BC MSP. Then, as above, Navigant
Consulting scaled product class 2, scaling the charging losses to account for any difference in
battery energy and re-weighting the performance in each mode by the average usage profile of
the applications in this product class.
The CSLs for this product class were based on some DOE tests and the distribution of the
scaled manufacturer data. The results are presented in below Table 5.74 and Table 5.77.
Table 5.76 Characteristics of typical units scaled for product class 8.
Rated
Battery
Energy

Rated
Battery
Voltage

Est.
Time in
Active

Est. 24Hour
Charge
Energy

Est.
ActiveOnly
System

Est.
Maint.
Power

Est.
NoBattery
Power

UEC

Wh

Wh

Eff.

kWh/yr

CSL

Example
Application

Power Tool

3.6

3.0

14.3

25%

0.3

1.5

Cell Phone

3.6

3.0

9.4

38%

0.2

0.9

Cell Phone

3.6

3.0

6.4

47%

0.1

0.6

Table 5.77 Resulting cost and efficiency relationship for BC product class 8.
MSP from
MSP from
UEC
Interviews Teardowns
CSL

kWh/yr

2009$

2009$

1.5

$0.62

0.9

$1.42

0.6

$2.17

5-152

5.6.14.4
Scaled Product Class 9: Low Energy, 12 V DC Input
This product class contains small consumer electronic products typically recharged using
the 12 volt output of an automotive cigarette lighter receptacle, in particular satellite navigation
units intended for use in cars. Through product disassembly, DOE noticed many similarities to
the design of BCs in product class 2, though, again, without the need for an EPS, as the 12 V DC
input is converted into the voltage necessary for battery charging and operation using a DC-DC
converter inside the unit. Using a similar approach to that for scaled product class 8, DOE
developed the results shown in Table 5.78 and Table 5.79 below.
Table 5.78 Characteristics of typical units scaled for product class 9.
Rated
Battery
Energy

Rated
Battery
Voltage

Est.
Time in
Active

Est. 24Hour
Charge
Energy

Est.
ActiveOnly
System

Est.
Maint.
Power

Est.
NoBattery
Power

UEC

Wh

Wh

Eff.

kWh/yr

CSL

Example
Application

Power Tool

3.6

3.0

36.5

19%

0.5

1.3

Cell Phone

3.6

3.0

23.3

27%

0.2

0.8

Cell Phone

3.6

3.0

15.4

38%

0.1

0.5

Table 5.79 Resulting cost and efficiency relationship for BC product class 9.
MSP from
MSP from
UEC
Interviews Teardowns
CSL

kWh/yr

2009$

2009$

1.3

$0.62

0.8

$1.11

0.5

$2.76

5-153

Environmental Protection Agency (EPA). ENERGY STAR Program Requirements for Products with Battery
Charging Systems: Eligibility Criteria. May 2007.
http://www.energystar.gov/ia/partners/product_specs/program_reqs/battery_chargers_prog_req.pdf
2
Ecos Consulting energy efficiency tests. Provided to DOE by Pacific Gas & Electric. July 7, 2009.

5-154

CHAPTER 6. PRODUCT PRICE DETERMINATION


TABLE OF CONTENTS
CHAPTER 6. PRODUCT PRICE DETERMINATION ................................................. 6-1

6.1
Introduction....................................................................................................... 6-1

6.2
Methodology..................................................................................................... 6-1

6.2.1
Key Assumptions .............................................................................................. 6-1

6.2.2
Role of End-Use Product Markups ................................................................... 6-2

6.2.3
Distribution Channels ....................................................................................... 6-2

6.2.4
Data Sources ..................................................................................................... 6-4

6.2.5
Baseline Markups.............................................................................................. 6-5

6.2.6
Incremental Markups ........................................................................................ 6-5

6.3
Markups by End-Use Product Category ........................................................... 6-6

6.4
Markups by Battery Charger and External Power Supply Product Class......... 6-7

6.5
Sales Taxes........................................................................................................ 6-9

LIST OF TABLES
TABLE 6.1 MARKUPS BY END-USE PRODUCT CATEGORY ............................... 6-7

TABLE 6.2 BASELINE MARKUP CALCULATION FOR EPS PRODUCT CLASS A3

.......................................................................................................................................... 6-8

TABLE 6.3 EXTERNAL POWER SUPPLY MARKUPS BY PRODUCT CLASS...... 6-8

TABLE 6.4 BATTERY CHARGER MARKUPS BY PRODUCT CLASS ................... 6-9

TABLE 6.5 STATE AND LOCAL SALES TAX RATES ........................................... 6-10

TABLE 6.6 WEIGHTED AVERAGE SALES TAX RATES BY CENSUS DIVISION....

........................................................................................................................................ 6-11

LIST OF FIGURES
FIGURE 6.1 PATHS OF DISTRIBUTION FOR BATTERY CHARGERS AND

EXTERNAL POWER SUPPLIES .................................................................................. 6-3

6-i

CHAPTER 6. PRODUCT PRICE DETERMINATION

6.1

INTRODUCTION

This chapter presents DOEs method for estimating the prices paid by end-use
consumers for baseline and more efficient battery chargers (BCs) and external power
supplies (EPSs), an analysis known as the Product Price Determination (PPD). The PPD
draws upon the market assessment, contained in chapter 3 of this technical support
document (TSD). The markups calculated in the PPD are necessary inputs to the lifecycle cost (LCC) and payback period (PBP) analyses, and the national impact analysis
(NIA).
The manufacturer selling price (MSP) is the price at which a BC/EPS
manufacturer sells a completed BC or EPS, usually to an end-use product manufacturer
(sometimes called an original equipment manufacturer or OEM)a. In the engineering
analysis, DOE estimated BC and EPS prices as MSPs for all product classes and
representative units at each candidate standard level (CSL). In the PPD, DOE derived two
kinds of markups, those that are applied to baseline costs (baseline markups) and those
that are applied to incremental cost increases due to standards (incremental markups).
Markups are applied to BCs and EPSs as they move through each step in the
distribution chain. The markup applied by an OEM (the manufacturer markup) and the
markup applied by an end-use consumer product retailer (the retail markup) can be
multiplied together to yield a composite markup. The final product prices, as estimated
using these markups and sales tax data, are used in the LCC and NIA to forecast the
increase in BC/EPS costs to the consumer that would result from standards.
6.2
6.2.1

METHODOLOGY
Key Assumptions

DOE made five key assumptions in conducting the PPD. They are explained in
the following subsections.
The dominant path to market establishes the retail price and, thus, the composite
markup for a product.
The markups applied to end-use products that use BCs and EPSs are proxies for
BC and EPS markups.
The baseline markups that manufacturers and retailers apply to end-use products
that use BCs and EPSs are equal to those companies average markups across
their entire product lines.

Expenses like labor and administrative costs remain fixed and need not be

recovered in the incremental markup. Profits and other operating costs are

assumed to be variable and to scale with the MSP.

For further discussion of the MSP and its derivation, see chapter 5 of this TSD.

6-1

Markups can be derived from inspection of companies public financial filings.


6.2.2

Role of End-Use Product Markups

The markups applied to end-use products are used as a proxy for BC and EPS
markups. BC and EPS markups generally cannot be measured directly, as these products
are typically components of, or accessories to, other end-use products. The specific
markups for BCs and EPSs at each stage are therefore assumed to be of equivalent
percentages to the markups applied to the end-use products they accompany. For
example, if a printer is marked up 30 percent by a retailer, then the individual
components of that printer, including the EPS, are assumed to be marked up 30 percent as
well.
6.2.3

Distribution Channels

Each company involved in manufacturing, distributing, and selling end-use


products applies a markup to cover business costs and maintain profit margins. To
determine which markups are applied to the MSP, DOE first needed to model the
distribution channels for BCs and EPSs. Figure 6.1 illustrates this model. The most
common path to market, as identified by DOE, is depicted by the gray arrows, while
alternative paths are depicted by the white arrows. Based on comments from the
Association of Home Appliance Manufacturers, DOE combined the distribution models
of BCs and EPSs. Despite minor variations, the two models are assumed to be similar in
their basic structure. (Pub. Mtg. Tr., No. 14 at pp. 229-231) DOE based the PPD analysis
and stages at which markups were applied on the most common path to market.

6-2

Component Manufacturer

BC/EPS Manufacturer

BC/EPS Distributor

End-Use Product Manufacturer (OEM)

Consumer Product Retailer/Distributor

Taxes

Consumer
Note that the widths of arrows and sizes of boxes are not drawn to scale and are not meant to be an exact
indication of a distribution paths relative prominence.
Figure 6.1 Paths of Distribution for Battery Chargers and External Power Supplies

BC and EPS distribution begins with component manufacturers, who produce the
circuitry, circuitry components, wiring, housing, and other materials needed to
manufacture BCs and EPSs. These are often sold directly to BC/EPS manufacturers, who
produce a finished BC or EPS, often for a specific end-use product manufacturer. DOE
identified some cases in which BCs or EPSs are manufactured directly by the OEMb.
BCs and EPSs are then typically purchased by an OEM, at the manufacturer
selling price, as an input to an end-use consumer product. The BC and/or EPS is typically
integrated into (or packaged with) a consumer product and marked up for sale to a retailer
or, less frequently, a wholesaler. This markup applied by the OEM will be referred to as
the manufacturer markup. While most consumer products are manufactured in an OEMowned factory, there is a trend towards the use of electronics manufacturing services
(EMS), which make consumer products under contract to multiple OEMs and take
advantage of economies of scale to source materials and components at lower costs. DOE
believes that products manufactured by EMSs are marked up similarly to those
manufactured in-house by OEMs.
Retailers also add a markup to the consumer products they sell. DOE has
identified a number of instances where the manufacturing and retail operations for a
product are owned and managed by one company. An example is Apple, Inc., which
manufactures consumer electronics that are sold in Apple-branded retail stores. DOE
b

For OEMs, it is more common to manufacture BCs in-house than it is to manufacture EPSs in-house.

6-3

believes that the markups on these products are similar to the combination of the
manufacturer and retailer markups found on other products, allowing the company to
maintain gross margins competitive with other companies in the industry.
It is the OEM and retailer markups that DOE applied to the MSP in this analysis
to determine the end-user product prices of BCs and EPSs. The MSP already takes into
account BC/EPS manufacturer markups. The majority of states and some local
governments then impose a sales tax, resulting in the final cost to consumers. Sales taxes
are discussed further in section 6.5.
A note on distributors
Some OEMs, particularly smaller manufacturers, opt to source components
through distributors for a variety of reasons, such as easier access to a wider array of
components. Distributors tend to have low margins and, due to high sales volumes, can
purchase and sell products at reduced prices. Distributors represent an additional step in
the chain; however, this step is uncommon. Given DOEs assumption that the most
common path to market sets the final product price, the presence of an EPS or BC
distributor in the distribution chain is assumed not to affect the final product price.
The distribution of EPSs and BCs for medical devices differs from those for other
devices, as many medical devices are prescribed by a doctor and are not available at
traditional retail outlets. These devices are therefore sold by medical distributors, rather
than retailers. DOE calculated the retail markups for medical devices using the financial
information from several large medical device distributors.
High-power EPSs used in amateur radio setups are typically sold as individual
components through small, specialized retailers. DOE assumes that these retailers
commonly purchase EPSs from EPS distributors rather than EPS manufacturers, given
their small purchasing volume. As such, the two markups that DOE applies to amateur
radios are a distributor markup (in lieu of an OEM markup) and a retailer markup.
6.2.4

Data Sources

Individual product markups are generally confidential and are not readily
available. As such, DOE used corporate sales revenue and the direct costs of products
sold, known as costs of goods sold (COGS), to estimate the average baseline markup
applied to all products that a company manufactures or sells. For each company analyzed,
DOE calculated and averaged markups for all fiscal years reported in the companys most
recent annual report (as far back as fiscal year 2006) to arrive at the average markup. The
average markup was then used to estimate the markups applied to the companys relevant
products that use BCs and/or EPSs. DOE sourced financial data from publicly available
filings with the Securities and Exchange Commission (SEC) for all domestic companies
analyzed, as well as for those foreign companies trading under an American Depositary
Receipt (ADR). For foreign public companies that trade only on foreign stock exchanges,
markups were calculated from annual reports and financial statements published on those
companies websites. DOE relied on these data under the assumption that financial data

6-4

reported by a publicly traded company in a quarterly or annual report have been verified
by an independent, certified auditor, and can therefore be considered accurate.
6.2.5

Baseline Markups

A markup is a percentage increase added to the input costs of a good or service so


that a company can cover its costs and earn a profit. Gross margin is the component of a
products price, added to COGS, that includes overhead costs (selling, general, and
administrative), research and development, other expenses, and profits. To calculate
baseline markups, DOE used the following equation:

MU BASE

1
1 GM

REV
COGS

EQ 1

where:

GM

REV COGS
REV

EQ 2

and:

MU BASE
GM
REV
COGS

=
=
=
=

baseline manufacturer or retailer markup,

corporate gross margin as a percentage of revenue,

revenue of the OEM/retailer, and

cost of goods/services manufactured or sold.

For example, if a computer manufacturer operates with revenues of $100 million


and has a COGS of $80 million, then the above calculations yield 1.25, or a markup of 25
percent applied to all computers manufactured. Therefore, it is assumed that any BCs or
EPSs integrated into the companys computer during the manufacturing process will be
marked up 25 percent.
6.2.6

Incremental Markups

DOE assumes a division of costs between those that do not scale with the MSP
(fixed costs) and those that do (variable costs). DOE used the baseline markups
(MUBASE), which cover all of a retailers costs (i.e., both fixed and variable costs), to
determine the sales price of baseline models. The composite baseline markup relates the
BC/EPS MSP to the final product price. DOE considers baseline models to be equipment
sold under existing market conditions (i.e., without new energy efficiency standards).
DOE calculated the baseline markups for manufacturers and retailers using Equation 1.
Incremental markups (MUINCR) are coefficients that relate the change in the
BC/EPS MSP of higher-efficiency models (incremental cost) to the change in the final
product price. Incremental markups are applied only to the incremental cost of these
higher efficiency models; the baseline markup is still applied to the baseline portion of
6-5

the MSP. Incremental markups cover only those costs that scale with a change in the
MSP (i.e., variable costs). The surveyed public financial filings did not typically separate
labor and occupancy costs from overall expenses, so DOE assumed these fixed costs to
be encompassed by selling, general, and administrative expenses (SG&A), which are
typically reported in financial statements. Incremental markups were calculated using the
following equation:

MU INCR

REV

COGS
COGS

SGA

REV SGA
COGS

EQ 3

where:

MU INCR
REV
COGS
SGA

=
=
=
=

incremental manufacturer or retailer markup,


revenue of the OEM/retailer, and
cost of goods/services manufactured or sold
selling, general, and administrative expenses

6.3

MARKUPS BY END-USE PRODUCT CATEGORY


To determine the markup applied to each BC and EPS product class (or
representative unit), DOE first assigned each application to one of 16 end-use product
categories, grouped by industry and similarity in manufacturing and/or retail practices. A
categorized list of end-use product applications is available in the Excel workbook that
accompanies this chapter. DOE then gathered gross margin data and calculated baseline
and incremental markups for the leading, publicly traded end-use product manufacturers
and retailersc in each category. DOE then calculated simple-average retailer and
manufacturer markups for each category and multiplied those two markups together to
obtain a composite markup for the category using equation 4:

MU COMP,i

MU MFG ,i

MU RET ,i

EQ 4

where:

MU RET ,i

= simple average retailer markup in category i,

MU MFG ,i

= simple average manufacturer markup in category i, and

MU COMP,i

= composite markup applied to BCs and EPSs that power applications in


category i

shows the retailer, manufacturer, and composite baseline and incremental


markups DOE calculated for each end-use product category.
Table 6.1

Determined by product surveys, research reports, and most popular products on top retail websites such as
www.amazon.com.

6-6

Table 6.1 Markups by End-Use Product Category


Manufacturer Markup
Retailer Markup
End-Use Product
Category
Baseline
Incremental
Baseline
Incremental

Composite Markup
Baseline

Incremental

Portable Audio &


1.42
1.15
1.48
1.16
2.11
1.34
Accessories
Mobile Telephony
1.52
1.29
1.41
1.13
2.14
1.46
Stationary
1.47
1.17
1.40
1.13
2.06
1.32
Telephony
Computers /
1.35
1.15
1.39
1.14
1.89
1.31
Accessories
Printers / MFDs
1.47
1.15
1.35
1.13
1.99
1.31
Geospatial
1.86
1.51
1.40
1.12
2.60
1.69
Equipment
Power Tools /
Outdoor
1.51
1.16
1.42
1.16
2.14
1.34
Appliancesd
Transport
1.55
1.20
1.49
1.14
2.31
1.37
Photo / Video
1.53
1.12
1.40
1.12
2.14
1.26
Floor Care
1.41
1.14
1.42
1.15
2.00
1.30
Games /
1.61
1.29
1.41
1.12
2.27
1.45
Entertainment
Personal Care
1.45
1.10
1.38
1.11
2.00
1.23
Medical
1.93
1.40
1.31
1.06
2.53
1.48
Home Systems
1.37
1.11
1.48
1.18
2.04
1.31
Amateur Radios
1.15*
1.03
1.47
1.13
1.69
1.16
Uninterruptible
1.35
1.15
1.49
1.17
2.01
1.35
Power Supply
Other**
1.48
1.18
1.47
1.13
2.17
1.33
* The first markup applied to amateur radios is an EPS distributor markup, rather than an OEM markup.
**Other contains applications that do not fit cleanly into any of the other 15 categories. Its markup was
calculated by multiplying the simple averages of all individual manufacturer and retailer markups.

6.4 MARKUPS BY BATTERY CHARGER AND EXTERNAL POWER SUPPLY


PRODUCT CLASS
In the engineering analysis, DOE identified 10 BC product classes and 15 EPS
product classes. Since, by design, each BC/EPS application can be found in only one of
the end-use product categories listed in Table 6.1, each BC/EPS end-use application is
associated with one composite markup. To calculate the markup for a product class (or, in
the case of EPS product class A1, one of four segments within that class), DOE
calculated the shipment-weighted average of the markups for applications associated with
that class. Table 6.2 gives an example of this.

Interviews with at least one manufacturer indicated that DOEs retail markup estimate is high and that a
markup of 1.13 may be more accurate. SEC data suggests that the retailer markup for power tools is 1.41. If
DOE identifies or receives further evidence supporting a lower markup for power tools or an alternate
manufacturer or retailer markup for any category, it will consider revising its estimates accordingly.

6-7

Table 6.2 Baseline Markup Calculation for EPS Product Class A3

EPS Product Class A3: AC-AC, Basic Voltage


Shipments as
Shipments in
Application
Percent of
Class A3
Product Class
Home Security Systems
4,219,000
52.8%
Aquarium Accessories
1,750,000
21.9%
Water Softeners/Purifiers
1,150,000
14.4%
Indoor Fountains
500,000
6.3%
Irrigation Timers
375,000
4.7%
Product Class Markup (weighted average markup)

Baseline
Markup
2.04
2.17
2.04
2.17
2.04
2.07

Table 6.3 and Table 6.4 display the results of these calculations for each BC and
EPS product class. A list of applications associated with each class can be found in the
market assessment (chapter 3) and the Excel workbook that accompanies the market
assessment.
Table 6.3 External Power Supply Markups by Product Class

Output

Class ID

Composite Composite
Baseline Incremental
Markup
Markup

Class A
A1: 0-10.25 W

2.16

1.40

A1: 10.25-39 W

2.00

1.31

A1: 39-90 W

2.00

1.34

A1: 90-250 W

1.89

1.31

Low Voltage

A2

2.19

1.47

Basic
Voltage

A3

2.07

1.31

Low Voltage

A4

2.17

1.33

100 W

X1

1.99

1.31

100 W

X2

2.27

1.45

>250 W
Basic
Voltage
Low Voltage
Basic
Voltage
Low Voltage

H1

1.69

1.16

M1

2.53

1.48

M2

2.53

1.48

M3

N/A*

N/A*

M4

N/A*

N/A*

Basic
Voltage

AC-DC

AC-AC

Non-Class A
Multiple-Voltage
High Power
For Medical
Devices

AC-DC

AC-AC

6-8

Output
For MotorOperated BC
Applications
and Detachable
Batteries

AC-DC

AC-AC

Composite Composite
Baseline Incremental
Markup
Markup

Class ID

Basic
Voltage
Low Voltage
Basic
Voltage
Low Voltage

B1

2.16

1.35

B2

2.01

1.23

B3

N/A*

N/A*

B4

N/A*

N/A*

* DOE did not identify any products in this class.


Table 6.4 Battery Charger Markups by Product Class

Battery
Energy

<100 Wh
ACDC
1003000
Wh

Class ID

Composite
Baseline
Markup

Composite
Incremental
Markup

Inductive
Connection

2.00

1.23

<4 V

2.13

1.40

410 V

2.17

1.33

>10 V

1.98

1.32

<20 V

2.32

1.37

20 V

2.28

1.36

2.31

1.37

<9 V Input

2.12

1.38

9 V Input

2.60

1.69

AC Output
from Battery

10

2.01

1.35

Battery
Voltage

>3000 Wh
DC-DC
AC-AC

6.5

SALES TAXES
A sales tax is a multiplicative factor applied to a products retail price that
increases the users first cost. DOE obtained information on State and local sales taxes
from the Sales Tax Clearinghouse. These data are displayed in Table 6.5 as weighted
averages that include county and city sales tax rates.

6-9

Table 6.5 State and Local Sales Tax Rates

State

Alabama

Combined
State and
Local Tax
Rate
(%)
8.15

Combined
State and
Local Tax
Rate
(%)
6.00

State

Kentucky

State

Combined
State and
Local Tax
Rate
(%)
5.80

Alaska
Arizona
Arkansas
California
Colorado
Connecticut

1.50
7.15
8.25
9.15
6.40
6.00

Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota

8.75
5.00
6.00
6.25
6.00
7.20

Delaware

0.00

Mississippi

7.00

Dist. Of
Columbia
Florida
Georgia
Hawaii
Idaho

6.00

Missouri

7.05

North
Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South
Carolina
South
Dakota
Tennessee

6.70
6.95
4.40
6.05

0.00
6.00
7.85
0.00

Texas
Utah
Vermont
Virginia

8.05
6.70
6.05
5.00

Illinois
Indiana

8.40
7.00

Montana
Nebraska
Nevada
New
Hampshire
New Jersey
New Mexico

6.95
6.40

8.75
6.00

Iowa
Kansas

6.85
7.00

New York
North
Carolina

8.45
7.80

Washington
West
Virginia`
Wisconsin
Wyoming

6.85
8.15
0.00
6.40
7.00
7.05
5.50
9.40

5.40
5.25

Source: Sales Tax Clearinghouse1

DOE then calculated average tax rates for each Census division and four large
States, weighted by 2008 state-level population. The population-weighted sales tax by
division is displayed in Table 6.6. Developing this distribution allowed DOE to correlate
the sales tax distribution with the electricity price distribution in the LCC. The table also
displays the national, population-weighted average sales tax that is used in the NIA,
where DOE did not use a distribution of inputs.

6-10

Table 6.6 Weighted Average Sales Tax Rates by Census Division

Division & Large


State Name

2008 Population

New England
14,429,720
Mid Atlantic
21,312,506
East North Central
46,500,668
West North Central
20,336,243
South Atlantic
40,657,961
East South Central
18,271,071
West South Central
11,068,576
Mountain
22,122,914
Pacific
12,483,503
New York State
19,541,453
California
36,961,664
Texas
24,782,302
Florida
18,537,969
U.S. Weighted Average:

PopulationWeighted Sales Tax


(%)
5.55
6.62
6.94
6.86
6.44
7.89
8.42
6.50
5.21
8.45
9.15
8.05
6.70
7.23%

Source: DOE analysis of data from the Sales Tax Clearinghouse and U.S. Census Bureau 2

6-11

REFERENCES
1

Sales Tax Clearinghouse, Inc. State sales tax rates along with combined average city and county rates.
2009. (Last accessed October 6, 2009). <https://thestc.com/STRates.stm>

U.S. Census Bureau. Annual Estimates of the Resident Population for the United States, Regions, States,
and Puerto Rico: April 1, 2000 to July 1, 2008, 2008. (Last Accessed October 6, 2009). Washington, DC.
<http://www.census.gov/popest/states/tables/NST-EST2008-01.xls >

6-12

CHAPTER 7. ENERGY USE ANALYSIS

TABLE OF CONTENTS

CHAPTER 7. ENERGY USE ANALYSIS................................................................................. 7-1

7.1
External Power Supplies ...................................................................................... 7-2

7.1.1 Modes and Application States ................................................................. 7-2

7.1.2 Loading Points ......................................................................................... 7-5

7.1.3 Calculating External Power Supply Energy Use ..................................... 7-6

7.1.4 EPS Power by Mode of Operation........................................................... 7-7

7.1.5 External Power Supply Usage Profiles .................................................. 7-10

7.1.6 External Power Supply Unit Energy Consumption Values ................... 7-11

7.2
Battery Chargers ................................................................................................ 7-15

7.2.1 Battery Charger Modes .......................................................................... 7-15

7.2.2 Battery Charger Usage Profiles ............................................................. 7-16

7.2.3 Calculating Battery Charger Energy Use............................................... 7-18

7.2.4 Battery Charger Unit Energy Consumption Values............................... 7-21

LIST OF TABLES
TABLE 7.1 SUMMARY OF EPS MODES ................................................................................ 7-2

TABLE 7.2 NOTEBOOK COMPUTER APPLICATION STATES AND LOADING POINTS ...

.......................................................................................................................................... 7-4

TABLE 7.3 LAN EQUIPMENT APPLICATION STATES AND LOADING POINTS........... 7-4

TABLE 7.4 PORTABLE DVD PLAYER APPLICATION STATES AND LOADING POINTS

.......................................................................................................................................... 7-4

TABLE 7.5 DEFAULT EPS APPLICATION STATES AND LOADING POINTS................. 7-5

TABLE 7.6 DEFAULT APPLICATION STATES AND LOADING POINTS FOR EPS USED

WITH BC......................................................................................................................... 7-5

TABLE 7.7 GENERIC EPS USAGE PROFILES..................................................................... 7-10

TABLE 7.8 GENERIC EPS USAGE PROFILE FOR A BC APPLICATION......................... 7-11

TABLE 7.9 EXTERNAL POWER SUPPLY REPRESENTATIVE UNIT AND ASSOCIATED

POWER OUTPUT RANGE .......................................................................................... 7-12

TABLE 7.10 EXTERNAL POWER SUPPLY UNIT ENERGY CONSUMPTION (KWH/YR) ...

........................................................................................................................................ 7-13

TABLE 7.11 PRODUCT CLASS A1 2.5 WATT REPRESENTATIVE UNIT .................... 7-13

TABLE 7.12 PRODUCT CLASS A1 18 WATT REPRESENTATIVE UNIT ..................... 7-14

TABLE 7.13 PRODUCT CLASS A1 60 WATT REPRESENTATIVE UNIT ..................... 7-15

TABLE 7.14 PRODUCT CLASS A1 120 WATT REPRESENTATIVE UNIT ................... 7-15

TABLE 7.15 GENERIC BATTERY CHARGER USAGE PROFILES ................................... 7-17

TABLE 7.16 BC SHIPMENTS-WEIGHTED AVERAGE USAGE PROFILES..................... 7-18

TABLE 7.17 MODE-SPECIFIC PERFORMANCE CHARACTERISTICS

MANUFACTURER DATA .......................................................................................... 7-23

TABLE 7.18 MODE-SPECIFIC PERFORMANCE CHARACTERISTICS TEST AND

TEARDOWN DATA..................................................................................................... 7-24

7-i

TABLE 7.19 BATTERY CHARGER UNIT ENERGY CONSUMPTION BY PRODUCT

CLASS BASED ON MANUFACTURER-SUPPLIED DATA (KWH/YR) ................ 7-25

TABLE 7.20 BATTERY CHARGER UNIT ENERGY CONSUMPTION BY PRODUCT

CLASS BASED ON TEST AND TEARDOWN DATA (KWH/YR) .......................... 7-25

TABLE 7.21 BATTERY CHARGER PRODUCT CLASS 1 BASED ON MANUFACTURERSUPPLIED DATA......................................................................................................... 7-26

TABLE 7.22 BATTERY CHARGER PRODUCT CLASS 2 BASED ON MANUFACTURERSUPPLIED DATA......................................................................................................... 7-27

TABLE 7.23 BATTERY CHARGER PRODUCT CLASS 2 BASED ON TEST &

TEARDOWN DATA..................................................................................................... 7-27

TABLE 7.24 BATTERY CHARGER PRODUCT CLASS 3 BASED ON MANUFACTURERSUPPLIED DATA......................................................................................................... 7-27

TABLE 7.25 BATTERY CHARGER PRODUCT CLASS 4 BASED ON MANUFACTURERSUPPLIED DATA......................................................................................................... 7-28

TABLE 7.26 BATTERY CHARGER PRODUCT CLASS 4 BASED ON TEST &

TEARDOWN DATA..................................................................................................... 7-28

TABLE 7.27 BATTERY CHARGER PRODUCT CLASS 5 BASED ON MANUFACTURERSUPPLIED DATA......................................................................................................... 7-29

TABLE 7.28 BATTERY CHARGER PRODUCT CLASS 5 BASED ON TEST &

TEARDOWN DATA..................................................................................................... 7-29

TABLE 7.29 BATTERY CHARGER PRODUCT CLASS 6 BASED ON MANUFACTURERSUPPLIED DATA......................................................................................................... 7-30

TABLE 7.30 BATTERY CHARGER PRODUCT CLASS 7 BASED ON MANUFACTURERSUPPLIED DATA......................................................................................................... 7-30

TABLE 7.31 BATTERY CHARGER PRODUCT CLASS 8 BASED ON MANUFACTURERSUPPLIED DATA......................................................................................................... 7-30

TABLE 7.32 BATTERY CHARGER PRODUCT CLASS 9 BASED ON MANUFACTURERSUPPLIED DATA......................................................................................................... 7-31

TABLE 7.33 BATTERY CHARGER PRODUCT CLASS 10 BASED ON

MANUFACTURER-SUPPLIED DATA ...................................................................... 7-31

TABLE 7.34 BATTERY CHARGER PRODUCT CLASS 10 BASED ON TEST &

TEARDOWN DATA..................................................................................................... 7-31

LIST OF FIGURES
FIGURE 7.1 CHARGING INPUT POWER OVER TIME (PORTABLE VIDEO GAME
CONSOLE)...................................................................................................................... 7-6

FIGURE 7.2 EXAMPLE MEASUREMENTS FROM AN EPS TEST PROCEDURE ............. 7-8

FIGURE 7.3 RELATIONSHIP BETWEEN EPS INPUT POWER AND OUTPUT POWER AT

FIVE LOADING POINTS .............................................................................................. 7-9

FIGURE 7.4 CALCULATING CHARGE ENERGY CONSUMPTION FROM 24-HOUR

ENERGY ....................................................................................................................... 7-19

FIGURE 7.5 ACTIVE AND MAINTENANCE TIME............................................................. 7-20

7-ii

CHAPTER 7. ENERGY USE ANALYSIS


The purpose of the energy use analysis is to identify how consumers use products and
equipment, and thereby determine the energy savings potential of energy-efficiency
improvements. For battery chargers (BCs) and external power supplies (EPSs), DOEs analysis
focuses on how end users operate BCs and EPSs with the consumer products they power.
It is in the energy use analysis that DOE derives unit energy consumption (UEC), which
is an input to the life-cycle cost (LCC) and payback period (PBP) analyses described in chapter
8. The LCC and PBP analyses require data on annual energy use because these data, along with
energy prices, establish the operating costs.
The engineering analysis described in chapter 5 reports energy use based on the DOE test
procedures. These tests provide standardized results that serve as the basis for comparing the
performance of different BCs and EPSs used under the same conditions. Actual usage in the field
varies depending on the conditions in which the appliances are operated.a The unit energy
consumption calculated in this chapter represents the typical annual energy consumption of a BC
or EPS in the field. A critical part of characterizing end-use loads for BCs and EPSs is
identifying usage profiles, which estimate the time a device spends in each mode in one year.
Because of the nature of BCs and EPSs, the usage profile of the device will be related to the
usage profile of the associated application. It is difficult to predict changes in usage, so DOE
assumes that usage profiles will not change over the analysis period.
BCs and EPSs are power conversion devices that transform input voltage to suitable
voltage for the end-use application or battery they are powering. To provide that output power,
EPSs and BCs consume power due to internal losses as well as overhead circuitry. The amount
of power the EPSs and BCs consume varies with the power demands of the load. Therefore, the
traditional method for calculating energy consumption by measuring the energy a product draws
from mains while performing its intended function(s) is not appropriate for BCs and EPSs.
Instead, energy consumption is taken to be the energy dissipated by the BC or EPS (losses) and
not delivered to the end-use product or battery. Once the energy and power requirements of those
end-use products and batteries have been determined, they are considered fixed, and DOE
considers only how standards would affect the energy consumption of BCs and EPSs themselves.
DOE used a single usage profile for each application to calculate unit energy
consumption for BCs and EPSs. However, usage varies by application and among users. For
some applications DOE developed an average usage profile based on different user types. The
following are some applications with usage profiles derived from more than one user type:
Digital cameras
Golf cars
Ink Jet Imaging Equipment
Notebook Computers
a

DOE estimated the power requirements and usage of various end-use applications that use BCs and EPSs based on
published reports, comments from interested parties, and test data. Unfortunately, some of the estimates were not
based on metered data but rather relied on power measurements and assumptions regarding usage. When data were
unavailable, DOE relied on its own estimates of power and usage.
7-1

Professional-grade power tools


Two-way radios
Video Game Consoles
DOE examined multiple usage profiles for applications where usage varies widely.
Typically, DOE examined a light user and a heavy user. Although user types vary significantly, a
heavy user might, for example, be found in the commercial sector, whereas light users might
often be found in the residential sector.
Section 7.1 explains how DOE calculated EPS energy consumption while section 7.2 is
devoted to BC energy consumption. There are two appendices associated with this chapter.
Appendix 7A shows application-level usage profiles, while Appendix 7B shows applicationlevel UECS.
7.1

EXTERNAL POWER SUPPLIES

This section describes EPS modes and application states, how UEC is calculated, and
results for each product class.
7.1.1 Modes and Application States
When describing usage and energy consumption from the perspective of the EPS, DOE
uses the term EPS mode. When describing usage and energy consumption from the perspective
of the application, DOE uses the term application state.
7.1.1.1 EPS Modes
An EPS can be in active mode, no-load mode, off mode, or unplugged. Table 7.1 gives a
summary of these modes, which are also discussed in chapter 3.
Table 7.1 Summary of EPS Modes
EPS Mode
Status of EPS
Connection to
Mains
Active
Connected
No-load
Connected
Off
Connected
Unplugged
Disconnected

Status of EPS
Connection to
Application
Connected
Disconnected
Disconnected

EPS On/Off Switch


Selection
(If Switch is Present)
On
On
Off

Active Mode: In active mode, the external power supply takes power from mains and
converts it to a form usable by the consumer product or load. Thus, in calculating usage profiles
and energy consumption, DOE considers active mode to include any condition where the EPS is
connected to both mains and the application.
No-Load Mode: EPCA defines no-load mode for EPSs as the mode of operation when an
external power supply is connected to the main electricity supply and the output is not connected
7-2

to a load. (42 U.S.C. 6291(36)(D)) DOE determined that for EPSs, no-load mode is equivalent to
standby, as explained in the Final Rule on Test Procedures for Battery Chargers and External
Power Supplies (Standby Mode and Off Mode), published in the Federal Register on March 27,
2009. (74 FR 13318)
Off Mode: Off mode is a mode applicable only to an EPS with an on/off switch in which
the EPS is connected to mains, is disconnected from the load, and the on/off switch is set to
off. This definition was promulgated in the final rule reference just above. Of the EPSs DOE
examined, only two Non-Class A high power units, which are used with amateur radios, included
on/off switches. In both cases, turning off the switch fully severed the circuit, creating a situation
electrically equivalent to the EPS being unplugged from mains. To estimate energy consumption,
DOE treated the time when the EPS switch is set to off as equivalent to unplugged time.
Unplugged: Unplugged simply means the EPS is disconnected from mains power. No
energy is consumed in this state.
7.1.1.2 Application States
All energy-consuming application states are part of active mode from the perspective of
the EPS. That is, since any energy-consuming application state requires the application to be
connected to the EPS, any energy-consuming application state is part of EPS active mode. The
number of states and power each requests varies by application.
DOE identified application states and loading points for each application. Loading points
are expressed relative to nameplate output power, which is the most power an EPS is capable of
delivering. DOE conducted loading point tests for three applications. Application states and the
corresponding loading points for these three applications are shown in Table 7.2, Table 7.3, and
Table 7.4. The average (across the three tested units) of the loading point corresponding to the
highest application state was 60 percent of nameplate output power. DOE used this loading point
as the default loading point in the operating application state. DOE assumed a 2 percent load
when the application is idle, and a load of 1 percent of nameplate output power when it is off or
asleep. The default loading points DOE used when test data were not available are shown in
Table 7.5 and Table 7.6.

7-3

Table 7.2 Notebook Computer Application States and Loading Points


EPS Mode Application State
Description

Charging the
Battery and
Operating
Operating - High

Active

Charging the
Battery
Operating

Sleep

Off

Device is charging the battery while


operating (no USB devices are attached).
(Sum of Operating and Charging the
Battery)
Device is on, battery is fully charged,
DVD drive is operating, USB devices
attached and powered.
Device is charging the battery while in
device is turned off.
Device is on, battery is fully charged,
DVD drive is not operating, no USB
devices attached.
Device is sleeping, battery is fully
charged, DVD drive is not operating, no
USB devices attached.
Device is off, battery is fully charged,
DVD drive is not operating, no USB
devices attached.

Percent of
Nameplate
Output Power
66%

60%

38%
28%

1.6%

0.6%

Source: Test results.

Table 7.3 LAN Equipment Application States and Loading Points


EPS Mode Application State
Description

Active

Operating Transmitting
Operating Not
Transmitting

Device is on and transmitting data.


Device is on but not transmitting data.

Percent of
Nameplate
Output Power
57%
54%

Source: Test results.

Table 7.4 Portable DVD Player Application States and Loading Points
EPS Mode Application State
Description

Operating - High
Active

Idle
Off

Device is on and operating, battery is fully


charged.
Device is on and idle, battery is fully
charged.
Device is off, battery is fully charged.

Source: Test results.

7-4

Percent of
Nameplate
Output Power
60%
54%
1%

Table 7.5 Default EPS Application States and Loading Points


EPS Mode Application State
Description

Operating - High
Active

Idle/Sleep
Off

Application is performing its intended


function.
Application is in idle, ready, or sleep
mode.
Application is turned off

Percent of
Nameplate
Output Power
60%
2%
1%

Source: DOE estimates.

Table 7.6 Default Application States and Loading Points for EPS used with BC
EPS Mode Application State
Description
Percent of
Nameplate
Output Power
Charging the
BC is charging the applications battery.
37%
Battery
Active
Maintenance
BC is in maintenance mode
17%
Source: DOE estimates.

7.1.2 Loading Points


For BC applications powered by an EPS, DOE examined test data to determine
appropriate loading points to characterize charging and maintenance application states. The
power requested by a BC varies considerably during charging, as illustrated by Figure 7.1. DOE
estimated average input power over the duration of a single charge by dividing estimated BC
active energy by estimated charge time. From this value, DOE deduced average EPS output
power during a charge using the tested average active-mode efficiency of the EPS. Given the
nameplate output power and average output power over a charge, DOE calculated the average
loading point over a charge. DOE test results indicated an average maintenance mode loading
point of 17 percent of nameplate output power. Application states and loading points for each
EPS application examined are in appendix 7A.

7-5

Input Power

4
3
2
1
0

Elapsed Time
Figure 7.1 Charging Input Power over Time (Portable Video Game Console)
7.1.3 Calculating External Power Supply Energy Use
EPS energy consumption (UEC) is the sum of energy consumed in active mode
(UECActive) and in no-load mode (UECNL).
UEC =

UECActive + UECNL

To describe EPS active-mode energy, UECActive, it is necessary to examine each activemode state:
UECActive =

UECActive1 + UECActive2 + UECActive3 +

Where UECActive(i) is EPS active-mode energy in application state i. EPS active-mode


energy in each application state is the product of EPS active power, PActive(i), and time, tActive(i) in
each state. Therefore,
UECActive(i) = PActive(i) tActive(i)
and
UECActive =

PActive1 tActive1 + PActive2 tActive2 +PActive3 tActive3 +

No-Load mode energy is given as:


UECNL =

PNo-Load tNo-Load

7-6

Where PNo-Load is no-load power and tNo-Load is time in no-load. Combining active and noload mode values provides total EPS unit energy consumption:
UEC =PActive1 tActive1 + PActive2 tActive2 +PActive3 tActive3 ++ PNo-Load tNoLoad

Thus, to calculate the energy consumption of an EPS, DOE combined the time values
(from usage profiles) with power values. Section 7.1.4 explains how these power values were
calculated, and section 7.1.5 explains how the time values were derived.
7.1.4 EPS Power by Mode of Operation
No-load mode power (PNo-Load) is the measured power drawn by the EPS from mains
while in no-load mode. Because the EPS is disconnected from the application, all of the power
drawn from mains is consumed by the EPS. For each candidate standard level (CSL), an
associated no-load mode power is given.
EPS power in active mode is a function of four factors: the nameplate output power of the
EPS, the proportion of full load required by the application (as discussed above), the active-mode
efficiency of the EPS, and no-load mode power. EPS power during active mode varies as the
power requirements of its load vary.
DOE used two different approaches to calculate EPS power in active modeone for
application states requiring 25 percent or more of the EPSs nameplate output power and another
for application states requiring less than 25 percent.
The approach for application states requiring 25 percent or more of the EPSs nameplate
output power is straightforward. The EPS test procedure measures the active-mode efficiency
(EPS) of the EPS at 25, 50, 75, and 100 percent of nameplate output power or current. The
active-mode efficiency of the EPS is then defined as the average of these four values (points 2, 3,
4, and 5 in Figure 7.2). As such, power is
PActive = POut (1/ EPS 1)
For example, an EPS delivering an output power of 10 watts at 75 percent efficiency
would consume:
PActive = 10 W (1/0.75 1) = 3.33 W

7-7

Input Power

3
X

2
X

1
(0W, PNL) X

0%

25%

50%

75%

100%

Output Power (Percent of Nameplate)

Figure 7.2 Example Measurements from an EPS Test Procedure


The approach for application states requiring less than 25 percent of the EPSs nameplate
output power blends measures of the EPSs active-mode efficiency and no-load mode power.
Most applications have some application states that require an output power below 25 percent of
nameplate. Additionally, many applications spend a significant portion of time in these states.
However, at these low power levels, the efficiency of an EPS is relatively low. Thus, DOE does
not use the active-mode efficiency measurement alone.
Instead, DOE calculates energy consumption in this region by interpolating between two
known points: (1) the no-load mode power and (2) the active-mode power at 25 percent of
nameplate output power. These are points 1 and 2 in Figure 7.3. At an output power of 25
percent, the energy consumed by the EPS is: PActive = POut (1/EPS 1). At an output power of
0 percent (no-load mode), EPS power is: P = PNL. Given these two known values, DOE assumes
that the energy consumed by an EPS at an output power between 0 and 25 percent of nameplate
output power must fall on a curve between these two points.

7-8

Input Power

3
X

2
X

1
(0W, PNL) X

0%

(25% x PNameplate, 25% x [PNameplate / ])

25%

50%

75%

100%

Output Power (Percent of Nameplate)

Figure 7.3 Relationship between EPS Input Power and Output Power at Five Loading
Points
DOE approximates this curve with a straight line between points 1 and 2, which can be
described by the following equation relating input power to output power:
PIn = m x POut + B
Where B is the y-axis intercept, or PNL, and m is the slope, or:
m = y/x

m = [(25% [PNameplate / EPS]) PNL] / [(25% PNameplate) 0]

Thus, for output power levels below 25 percent of nameplate output power, the input
power of the EPS is:
PIn = {[(25% [PNameplate / EPS]) PNL] / [25% PNameplate]} POut + PNL
Since

PActive = PIn POut

Then
PActive = {[(25% [PNameplate / EPS]) PNL] / [25% PNameplate]} POut + PNL POut
Or
PActive = ({[(25% [PNameplate / EPS]) PNL] / [25% PNameplate]}-1) POut + PNL

7-9

7.1.5 External Power Supply Usage Profiles


Because usage of an EPS is tied to usage of the application, DOE identified usage
profiles for a variety of applications. Usage profiles for many of the most common applications
were developed based on published research and stakeholder comments. Notable exceptions
include computer speakers and portable video game systems. Where usage data were lacking,
DOE assigned the application a generic usage profile. The five generic EPS usage profiles DOE
used in the preliminary analysis are shown in Table 7.7. For EPSs that power BC applications,
DOE assigned a usage profile that is consistent with the associated BC profile. An example of
this is shown in Table 7.8. All usage profiles and sources are shown in appendix 7B.
DOE used a single usage profile for each application to calculate unit energy
consumption. For most applications the usage profile represents a typical user. However, usage
can vary significantly depending on the user. Therefore, for some applications, including
notebook computers and video game consoles, DOE developed multiple usage profiles to
account for different usage patterns. DOE also examined usage profiles specific to commercial
users for digital cameras, golf cars, professional-grade power tools, and two-way radios. DOE
then calculated a weighted-average usage profile for each of these applications based on an
estimated distribution of user types. It is these weighted-average usage profiles that were used in
subsequent analyses.
Table 7.7 Generic EPS Usage Profiles
INFREQUENTLY USED (APPLICATION IS MOSTLY TURNED OFF)
EPS
Mode

Application State

Operating
Active
Idle
Off
No-load
Disconnected from
EPS
Unplugged Disconnected from
EPS

Percent of Nameplate Output


Power
60%
2%
1%
-

Time per Week


(hours)
5
3
160
0

INFREQUENTLY USED (EPS IS MOSTLY UNPLUGGED)


EPS
Mode

Application State

Operating
Active
Idle
Off
No-load
Disconnected from
EPS
Unplugged Disconnected from
EPS

Percent of Nameplate Output


Power
60%
2%
1%
-

Time per Week


(hours)
5
3
0
0

160

7-10

ALWAYS ACTIVE (APPLICATION IS ALWAYS PLUGGED IN AND OPERATING)


EPS
Mode
Active
No-load

Application State

Operating
Disconnected from
EPS
Unplugged Disconnected from
EPS

Percent of Nameplate Output


Power
60%
-

Time per Week


(hours)
168
0

MOSTLY NO-LOAD
EPS
Mode

Application State

Operating
Idle
No-load
Disconnected from
EPS
Unplugged Disconnected from
EPS
Active

Percent of Nameplate Output


Power
60%
2%
-

Time per Week


(hours)
21
21
126

MOSTLY NO-LOAD AND UNPLUGGED


EPS
Mode

Application State

Operating
Idle
No-load
Disconnected from
EPS
Unplugged Disconnected from
EPS
Active

Percent of Nameplate Output


Power
60%
2%
-

Time per Week


(hours)
21
21
63

63

Table 7.8 Generic EPS Usage Profile for a BC Application


[hours/day]
BC
BC
No
BC Usage Profile
Unplugged
Active Maintenance Battery
2 Charges per week

1.4

6.1

EPS Usage Profile

Active1

Active2

2 Charges per week

1.4

6.1

8.5

No-load Unplugged
8.5

[#/day]
Off

Charges

2/7

7.1.6 External Power Supply Unit Energy Consumption Values


DOE calculated UEC at each CSL for EPSs associated with each application. These
application-level UECs are inputs to the LCC analysis, which is conducted on each
representative unit.
7-11

For each application, DOE selected an EPS nameplate output power that was typical
based on a market survey of common applications (see BCEPS_Master_Survey.xls for more
details). Based on this information, DOE then assigned each application to the representative unit
that best matched the nameplate output power of the applications EPS. Table 7.9 shows how
applications were grouped based on nameplate output power. Some applications are common to
more than one representative unit. Some notebook computers, for example, are shipped with 120
watt EPSs while others are shipped with 60 watt EPSs. Applications with wide ranging EPS
nameplate output powers were split into two or more groups as appropriate.
Table 7.9 External Power Supply Representative Unit and associated Power Output range
Representative Nameplate Output Nameplate
Range of
Unit
Power
Output Voltage
Nameplate Output Powers
[W]

[V]

[W]

2.5

0-10.25

18

12

10.26-39

60

15

40-90

120

19

91-250

Once EPSs were sorted into these four groups, DOE calculated UEC values of each
application using the nameplate output power of the representative unit based on the usage
profile and loading points specific to the application. In this way, DOE could ensure that the
LCC analysis weighed the incremental costs for an EPS of a given output power with the energy
cost savings for an EPS with that same output power. Because the EPS LCC samples units at the
application level, this methodology further ensures that each applications unique usage profile
and loading points are considered in the calculation. The sections that follow examine each
representative unit in more detail.
DOE also calculated weighted-average UECs for each unit of analysis in the NIA.
Application shipments were used as weights. Table 7.10 shows these shipment-weighted average
UECs for AC-DC basic voltage EPSs (product class A1) and for product classes A2, A3, and A4.
DOE did not develop cost estimates for product classes A2, A3, and A4 (the scaled product
classes) and instead borrowed cost and efficiency data from product class A1 for these classes.
Specifically, DOE assumed that all products in A2 and A4 had nameplate output power of 2.5 W
while products in A3 had nameplate output power of 18 W. In this way, DOE was able to ensure
an appropriate cost/efficiency relationship in the NIA.

7-12

Table 7.10 External Power Supply Unit Energy Consumption (kWh/yr)


Nameplate
Class ID Output Power
CSL 0
CSL 1
CSL 2
CSL 3
(W)

CSL 4

0-10.25

3.8

2.5

1.9

1.7

1.6

10.26-39

15.5

11.8

9.7

7.9

6.6

40-90

11.7

10.3

8.7

7.3

5.4

91-250

19.4

16.8

14.2

13.6

9.4

A2

All

2.4

1.5

0.8

0.7

A3

All

15.7

12.0

9.9

8.0

6.7

A4

All

8.8

5.8

5.0

4.4

4.3

A1

7.1.6.1 Product Class A1: Basic Voltage AC/DC Class A External Power Supplies
Product class A1 is the largest EPS product class with over 200 million units shipped in
2008. Four representative units, with nameplate output powers of 2.5, 18, 60, and 120 W were
identified and analyzed for this product class. Effectively, this class has been subdivided into
four segments, and every EPS in this class has been assigned to one of the four segments based
on its nameplate output power.
Cordless phones and answering machines (which DOE assumes also have a cordless
phone) make up the majority of units in the segment associated with the 2.5 watt representative
unit. Cordless phone and answering device EPSs are always in EPS active mode. UECs for each
of the top five applications in this segment are shown in Table 7.11.
Table 7.11 Product Class A1 2.5 Watt Representative Unit
Shipments
CSL 0
CSL 1
CSL 2
Top Applications by
Shipments
(Units)
(kWh/year)

CSL 3

CSL 4

Answering Machines

20,175,000

5.0

3.3

2.8

2.5

2.4

Cordless Phones

19,151,000

5.0

3.3

2.8

2.5

2.4

Mobile Phones

13,140,000

2.6

1.6

1.1

0.9

0.8

Portable Video Game


Systems

10,884,000

2.9

1.8

0.7

0.4

0.1

0.1

1.7

1.6

0.8

In-Vehicle GPS

7,660,000

0.2

0.1

Other Applications

5,217,000

Weighted Average

3.8

2.5

0.1
1.9

Notes: Includes all AC/DC basic voltage Class A EPSs with nameplate output power up to 10.25 watts.

LAN equipment makes up the majority of units in the segment associated with the
18 watt representative unit. Because LAN equipment is always on, energy use is also relatively
7-13

high. DOE performed loading point tests on two pieces of LAN equipment. Both units consumed
the same amount of energy whether they were transmitting or idle. UECs for each of the top five
applications in this segment are shown in Table 7.12.
Table 7.12 Product Class A1 18 Watt Representative Unit
Top Applications by
Shipments
CSL 0
CSL 1
CSL 2
Shipments
(Units)
(kWh/year)
LAN Equipment

CSL 3

CSL 4

27,581,000

22.6

17.6

14.7

12.2

10.3

VoIP Adapters

8,845,000

6.1

4.0

3.0

2.0

1.5

Digital Picture
Frames

7,472,000

29.9

23.2

16.1

13.6

Portable DVD
Players

7,140,000

0.5

0.4

MP3 Speaker Docks

7,012,000

5.8

3.8

1.8

1.3

Other Applications

8,671,000

Weighted Average

15.5

11.8

7.9

6.6

19.4
1

0.7
0.6
2.8
9.7

Notes: Includes all AC/DC basic voltage Class A EPSs with nameplate output power between 10.25 and 39 watts.

A majority of EPSs power notebook computers in both the third and fourth segments,
which are associated with the 60 and 120 watt representative units, respectively. DOE considered
two user types for notebook computers based on a report by the Natural Resources Defense
Council.1 The road warrior usage profile has significant unplugged time whereas the desktop
replacement profile has very little unplugged time and more time in operating modes. DOE
incorporated another active-mode state into the notebook profiles by incorporating time playing a
DVD.2
Video game consoles are the second most common application in the third segment. DOE
defined two usage profiles for this application, one for a light user and one for a heavy user. The
usage profiles were based on in-home usage audits of video game consoles conducted by The
Nielsen Company in 2006. 3 DOE assumed 80 percent of users are light users and 20 percent are
heavy users. DOE also incorporated DVD usage for video game consoles. DOE estimated that
DVD usage did not vary among user types, and that one-third of video game consoles would be
used as a DVD player. As part of the determination analysis, DOE identified loading points for
the Xbox 360, which uses a non-Class A EPS. DOE applied these same loading points to video
game consoles with Class A EPSs. UECs for the four applications in the third segment and the
two applications in the fourth segment are shown in Table 7.13 and Table 7.14, respectively.

7-14

Table 7.13 Product Class A1 60 Watt Representative Unit


Shipments
CSL 0
CSL 1
CSL 2
Top Applications by
Shipments
(Units)
(kWh/year)
Notebooks

30,225,000

10.9

9.7

Video Game
Consoles

13,512,000

13.4

11.9

8.1

CSL 3

CSL 4

6.8

5.0

8.3

6.1

9.1

6.6

9.9

Ink Jet Imaging


Equipment

9,339,000

Netbooks

3,700,000

8.9

7.9

6.5

5.4

3.9

11.7

10.3

8.7

7.3

5.4

Weighted Average

11.2

12.0
10.5

Notes: Includes all AC/DC basic voltage Class A EPSs with nameplate output power between 40 and 89 watts.

Table 7.14 Product Class A1 120 Watt Representative Unit


Shipments
CSL 0
CSL 1
CSL 2
Top Applications by
Shipments
(Units)
(kWh/year)
Notebooks
LED Monitors
Weighted Average

CSL 3

CSL 4

10,075,000

19.3

16.7

14.1

13.4

9.4

160,000

28.6

24.7

21.1

20.1

14.0

19.4

16.8

14.2

13.6

9.4

Notes: Includes all AC/DC basic voltage Class A EPSs with nameplate output between 90 and 250 watts.

7.2

BATTERY CHARGERS

This section describes battery charger modes, how UEC is calculated, and results for each
product class.
7.2.1 Battery Charger Modes
For the purposes of calculating energy consumption, DOE considers a BC to always be in
one of the following five states, which are also discussed in chapter 3:
Active Mode: The DOE test procedure for BCs defines active mode as the condition
in which the battery charger is connected to a power source; a battery is attached to
the charger; and the battery charger is charging a depleted battery, equalizing its cells,
or, performing other one-time or limited-time functions necessary for bringing the
battery to the fully charged state. (42 U.S.C. 6295(gg)(1)(A)(i))
Maintenance Mode: The DOE test procedure for BCs defines maintenance mode as
the condition in which the battery charger is connected to mains power and the
battery is fully charged, but is still connected to the charger.
Standby Mode or No-Battery Mode: The DOE test procedure for BCs defines nobattery mode as the condition in which the battery charger is connected to mains
7-15

power and no battery is attached to the charger. DOE revised its BC test procedure to
define standby mode as equivalent to no-battery mode for BCs. (74 FR 13318)
Off Mode: For BCs, off mode is the condition in which the charger is connected to
mains power, the charger is not connected to the battery, and all switches on the
device are in the off position.
Unplugged: The battery charger is disconnected from mains power. No energy is
consumed in this mode.
7.2.2 Battery Charger Usage Profiles
The BC usage profiles DOE used in the preliminary analysis made use of the following
measures:
(1) Equivalent Charges per Day: The number of full charges completed in a day. This is
the product of number of the frequency of charging (charges per day divided and days
per week) and the depth of discharge of each charge. An application charged once per
week from 100 percent depth of discharge would have 1/7 charges per day.
(2) Total Time in Active and Maintenance Mode: The sum of time spent in active and
maintenance modes over 24 hours.
(3) Time in No-Battery (Standby) Mode: The time per day spent in no-battery mode.
(4) Time in Off Mode: The time per day spent in off mode.
(5) Time Unplugged: The time per day spent unplugged.
Because usage of a BC is tied to usage of the application, DOE gathered usage profiles
for a variety of applications. DOE received a number of BC-specific usage profiles from PG&E
which it incorporated into its analysis. (DOCKET # 30) Where usage data were lacking, DOE
assigned the application a generic usage profile. The generic BC usage profiles DOE used in the
preliminary analysis are shown in Table 7.15.
Application usage depends strongly on the individual user. For most applications, DOE
assigned a single usage profile to represent all users. For some applications, DOE developed
multiple usage profiles to account for different users. DOE then calculated a weighted-average
usage profile based on an estimated distribution of user types. Usage profiles for each application
are detailed in appendix 7A.

7-16

Table 7.15 Generic Battery Charger Usage Profiles


No
Active +
Battery
Unplugged
Maintenance
(Standby)
Description

Off

Charges
Source
Number
per Day

Hours per Day

PG&E for
1/50 emergency
backup systems
Based on PG&E
1/7 for electric
housewares

All
Maintenance

24.0

Mostly
Maintenance

23.5

0.5

1.0

23.0

1/7 DOE estimate

7.5

8.5

8.0

2/7

6.0

7.0

11.0

21.8

2.2

12.0

12.0

1 Charge per
week (mostly
unplugged)
2 Charges per
week
5 Charges per
week
7 Charges per
week - light
use
7 Charges per
week - heavy
use

PG&E for MP3


player
PG&E for
5/7
mobile phone
Based on PG&E
1 for cordless
phone
1

PG&E for
wheelchair

Application usage profiles were then weighted by application shipments to calculate a


shipments-weighted average usage profile for each product class, shown in Table 7.16.

7-17

Table 7.16 BC Shipments-Weighted Average Usage Profiles


BC Product
Active +
No Battery
Unplugged
Class
Maintenance
(Standby)

Off

Charges

Hours per Day


1
2
3
4
5
6
7
8
9
10

23.9
9.7
5.6
16.8
7.7
15.4
7.7
6.5
1.1
24.0

0.1
5.0
0.2
0.3
0.5
8.6
8.1
7.5
0.1
0

0
9.4
18.1
6.9
15.8
0
8.1
10.1
22.8
0

0
0
0.1
0.1
0
0
0
0
0
0

Number per
Day
0.26
0.58
0.22
0.88
0.55
0.46
0.36
0.43
0.15
0

Note: Usage times may not sum to exactly 24 hours due to rounding.

7.2.3 Calculating Battery Charger Energy Use


7.2.3.1 General Energy Use Methodology
UEC represents the annual energy consumption of a battery charger. To accurately
represent usage in the field, DOE calculates UEC by combining a usage profile with the energy
performance characteristics for each energy-consuming mode. For BCs, UEC over a given time
period is the sum of:
(1) Charge Energy Consumption over time: The product of the number of 24-hour charge
cycles and the 24-hour charge cycle energy consumption measured in the test
procedure, less battery energy;
(2) Maintenance Mode Energy Consumption: The product of the time spent in
maintenance mode and the power drawn while in maintenance mode;
(3) No Battery (Standby) Mode Energy Consumption: The product of the time spent in
standby mode and the power drawn while in standby mode; and
(4) Off Mode Energy Consumption: The product of the time spent in off mode and the
power drawn while in off mode.
Charge Energy Consumption: Given the 24-hour charge and maintenance energy
measurement at each CSL from the test procedure, E24, DOE calculated charge energy
consumption, ECharge, by subtracting battery energy, EBatt and excess maintenance energy, over
the measurement period. Excess maintenance energy is calculated as maintenance power, PMaint,
multiplied by time beyond that spent actively charging, tCharge. Therefore:
7-18

ECharge = E24 EBatt (PMaint (24 tCharge))


An example ECharge calculation is illustrated in Figure 7.4. In this case, E24 is represented
by the area under the blue curve, and includes the energy lost in the BC during charge, as well as
the battery energy and any excess maintenance mode losses. The excess maintenance mode
energy, represented by the area under the red curve, is (PMaint (24 tCharge)), or in this example,
approximately 1.2 W (24 h 7.5 h) = 19.8 Wh.

(W)

Charge and Maintenance


Mode Power
Maintenance Mode Power
Charge Energy

EBatt

12

15

18

21

24

Figure 7.4 Calculating Charge Energy Consumption from 24-Hour Energy


Charge energy over the course of a day, EChargeDay is the product of ECharge and the number
of charges per day, n:
EChargeDay = ECharge n
Maintenance Mode Energy Consumption: Maintenance mode energy, EMaint, is calculated
by multiplying PMaint by the daily amount of time the BC is plugged in and attached to the
battery, tActive + tMaint , but not charging it. The time the BC is plugged in and attached to the
battery and the charges per day are determined by the usage profile, while the time per charge,
tCharge, is dictated by the charge rate of the BC itself:
EMaint = PMaint (tActive + tMaint (tCharge n))
No Battery (Standby) Mode Energy Consumption: No battery (standby) energy, ENoBatt is
simply:
ENoBatt = PNoBatt tNoBatt
Off Mode Energy Consumption: Likewise, off-mode energy, EOff is:

7-19

EOff = POff tOff


Given energy consumption in each mode, the complete annual unit energy consumption
calculation is represented as:
UEC = 365 ([n (E24 EBatt PMaint (24 tCharge))] + [PMaint (tActive + tMaint
(tCharge n)] + [PNoBatt tNoBatt] + [POff tOff])
Or more simply,
UEC = 365 (EChargeDay + EMaint + ENoBatt + EOff)
7.2.3.1 Modified Energy Use Methodology for Some Slow Chargers
In some cases the methodology described above had to be modified because the
performance of the BC at a particular CSL did not fit with the application-specific usage profile.
For example, mobile phones (product class 2) typically use fast chargers and their usage profile
reflects a fast charging BC. However, CSL 0 in product class 2 is based on a slow charger
typically used with small tools and cordless phones. The mobile phone usage profile does not
allow enough time for the BC at CSL 0 to make a complete charge.
Because each CSL has an underlying charge time associated with it, a problem arises if it
will take longer to charge the battery than the usage profile will allow. That is,
tChargeCSL n > tActive + tMaint.
This problem is illustrated by Figure 7.5. BCs at CSLs 1, 2, and 3 complete their charges
within the time allotted by the usage profile whereas a BC at CSL 0 takes additional time to
complete the charge.
Time Plugged In and Connected to Battery
Charging

Maintenance

CSL3
CSL2
CSL1
CSL0
Figure 7.5 Active and Maintenance Time
In order to correctly account for charge energy consumption, it is necessary to allow the
BC enough time to complete the charge. To account for the extra time needed, DOE allowed the
7-20

BC to complete its charge by reducing the time spent unplugged or in no battery mode. The
resultant energy consumption in this case is calculated as follows.
Charge Energy Consumption (as before):
ECharge = E24 EBatt (PMaint (24 tCharge))
EChargeDay = ECharge n
Maintenance Mode Energy Consumption: Again, this modified methodology was used in
cases when the charge time of some slow BCs exceeded the time allocated under the given usage
profile for both active and maintenance modes (tActive + tMaint). No time remained for
maintenance mode, resulting in an EMaint of zero.
If PMaint ((tActive + tMaint) (tCharge n)) 0

Then EMaint = 0

No Battery (Standby) Mode Energy Consumption: If the charge time of a slow BC


exceeds its allocated time in active and maintenance modes, that excess time must be accounted
for. DOE deducted the extra time needed to complete a charge from the time spent unplugged
(with zero energy consumption). If the unplugged time provided by the usage profile was
insufficient to make up the excess, DOE deducted the time from the no-battery time.
If tChargeCSL n + tNoBatt 24 h

Then ENoBatt = PNoBatt tNoBatt

If tChargeCSL n + tNoBatt > 24 h

Then ENoBatt = PNoBatt (24 h - tChargeCSL n)

Since none of BC usage profiles spent any time in off mode, DOE did not adjust the off
time.
Off Mode Energy Consumption (as before):
EOff = POff tOff
Although the use of the modified methodology required a case-by-case modification of
the usage profile, DOE considers this appropriate because usage profiles represent only the
average or typical use of a BC with a given application. There will be departures from the usage
profiles, and using the above methodology DOE was able extend its analytical framework to the
less typical uses while continuing to model expected user behaviori.e., if the charger takes
longer to charge than expected, the user will leave it plugged in longer.
7.2.4 Battery Charger Unit Energy Consumption Values
DOE expresses each CSL in terms of annual unit energy consumption based on the weighted
average usage profiles described in section 7.2.2. In order to accurately represent the variety of
7-21

usage that occurs in the field, it is necessary to combine each application-specific usage profile
with mode-specific performance characteristics. DOE took the UECs and the mode-specific
performance characteristics - E24, EBatt, PMaint , tCharge, PNoBatt, and POff, presented in chapter 5.
These values can be found in Table 7.17 and Table 7.18. DOE then combined these
characteristics with application-level usage to calculate application-level UECs.

7-22

Table 7.17 Mode-Specific Performance Characteristics Manufacturer Data


NoRated
24-Hour
Maint.
Est. Time
Battery
Off Mode
Product
Battery
Charge
Mode
CSL
in Active
Mode
Power
Class
Energy
Energy
Power
(h)
Power
(W)
(Wh)
(Wh)
(W)
(W)
0
1.5
26.7
24.0
1.2
0.5
0.0
1
1.5
19.3
24.0
0.8
0.4
0.0
1
2
1.5
10.8
24.0
0.4
0.2
0.0
3
1.5
5.9
24.0
0.2
0.1
0.0
0
3.0
41.1
22.0
1.8
0.7
0.0
2
1
3.0
31.3
14.5
1.1
0.4
0.0
2
3.0
9.8
6.9
0.4
0.1
0.0
0
10.0
91.3
16.1
4.7
1.0
0.0
3
1
10.0
65.7
10.3
2.5
0.6
0.0
2
10.0
21.4
1.1
0.2
0.2
0.0
0
20.0
142.9
13.0
3.4
1.2
0.0
4
1
20.0
57.4
1.5
1.0
0.5
0.0
2
20.0
42.5
1.0
0.7
0.2
0.0
0
800.0
1830.0
13.0
10.0
10.0
0.0
1
800.0
1620.0
12.0
10.0
10.0
0.0
5
2
800.0
1301.1
8.0
0.1
0.1
0.0
3
800.0
1175.1
8.0
0.0
0.0
0.0
0
384.0
720.8
20.0
5.0
1.2
0.0
1
384.0
620.0
16.0
0.5
0.5
0.0
6
2
384.0
524.6
16.0
0.3
0.3
0.0
3
384.0
491.3
8.0
0.0
0.0
0.0
0
3750.0
5904.7
8.0
3.6
3.7
0.0
1
3750.0
5607.5
8.0
3.3
3.2
0.0
7
2
3750.0
5235.9
8.0
2.9
2.6
0.0
3
3750.0
4864.3
8.0
2.6
2.0
0.0
0
2.0
14.3
3.0
0.3
0.0
8
1
2.0
9.4
3.0
0.2
0.0
2
2.0
6.4
3.0
0.1
0.0
0
3.6
36.5
3.0
0.5
0.0
9
1
3.6
23.3
3.0
0.2
0.0
2
3.6
15.4
3.0
0.1
0.0
0
70.0
0
24.0
2.2
0.0
1
70.0
0
24.0
0.7
0.0
10
2
70.0
0
24.0
0.5
0.0
3
70.0
0
24.0
0.2
0.0

7-23

Table 7.18 Mode-Specific Performance Characteristics Test and Teardown Data


No24-Hour
Maint.
Battery
Est. Time
Battery
Off Mode
Product
Charge
Mode
CSL
Energy
in Active
Mode
Power
Class
Energy
Power
(Wh)
(h)
Power
(W)
(Wh)
(W)
(W)
1
2.9
33.5
21.4
1.4
0.3
0.0
2
2
2.8
8.8
2.6
0.1
0.1
0.0
0
16.3
167.5
0.6
5.9
2.2
0.0
4
1
13.4
52.6
0.9
1.4
1.4
0.0
2
22.9
44.6
2.7
0.6
0.1
0.0
0
830.7
2036.9
12.8
21.2
20.1
0.0
5
1
762.1
1647.3
15.9
11.9
11.6
0.0
0
78.0
149.9
24.0
2.2
N/A
0.0
10
1
51.7
93.0
17.3
0.7
N/A
0.0
DOE calculated UEC at each CSL for each application in each product class. These
application-level UECs are inputs to the LCC. Then, for each product class, DOE calculated the
shipment-weighted average of the application-level UECs. These UECs are used in the NIA.
DOE derived one set of UECs from manufacturer-supplied cost-efficiency data and another from
test & teardown data. The former are shown in Table 7.19 and the latter in Table 7.20. The
sections that follow examine each product class in more detail and include the top applicationlevel UECs. UECs for all applications can be found in appendix 7B.

7-24

Table 7.19 Battery Charger Unit Energy Consumption by Product Class based on
Manufacturer-Supplied Data (kWh/yr)
CSL 0
CSL 1
CSL 2
CSL 3
Battery
Battery
Class
Energy
Voltage
ID
(kWh/yr)
(kWh/yr)
(kWh/yr)
(kWh/yr)

<100
Wh
ACDC

100
3000
Wh
>3000
Wh
DC-DC

AC-AC

Inductive
Connection

10.0

7.0

3.5

1.5

<4 V

10.0

6.0

1.2

4<10 V

10.0

5.4

1.0

10 V

39.4

10.5

6.1

<20 V

202.3

158.4

100

75.0

20 V

69.3

41.9

25.0

18.0

All

290

250

200

150

<9 V Input

1.5

0.9

0.6

9 V Input

1.3

0.8

0.5

AC Output
from
Battery

10

19.6

6.4

4.0

1.5

Table 7.20 Battery Charger Unit Energy Consumption by Product Class based on Test and
Teardown Data (kWh/yr)
CSL 0
CSL 1
CSL 2
CSL 3
Battery
Battery
Class
Energy
Voltage
ID
(kWh/yr)
(kWh/yr)
(kWh/yr)
(kWh/yr)

<100
Wh
ACDC

100
3000
Wh
>3000
Wh
DC-DC

AC-AC

Inductive
Connection

<4 V

7.4

1.2

4<10 V

10 V

39.4

10.5

6.1

<20 V

232.7

179.3

20 V

All

<9 V Input

9 V Input

AC Output
from
Battery

10

19.6

6.4

7-25

7.2.4.1 Battery Charger Product Class 1: Inductive Connection, <100 Wh


This category includes rechargeable toothbrushes and water jets. DOE applied the same
usage profile to each application. These products employ a cradle charger and are always
plugged in. UECs for each of the two applications in this product class are shown in Table 7.21.
Table 7.21 Battery Charger Product Class 1 based on Manufacturer-Supplied Data
Shipments
CSL0
CSL1
CSL2
CSL3
Top Applications by
Shipments
(Units)
(kWh/year)
(kWh/year) (kWh/year) (kWh/year)
Rechargeable
4,868,000
10.0
7.0
3.5
1.5
Toothbrushes
Rechargeable Water
487,000
10.0
7.0
3.5
1.5
Jets
Weighted Average
10.0
7.0
3.5
1.5
Notes: Assumes a battery voltage of 1.2 V and battery energy of 0.8 Wh.

7-26

7.2.4.2 Battery Charger Product Class 2: <4 V Battery, <100 Wh


Mobile phones dominate product class 2 with over 100 million unit shipments. DOE
based its usage profile for mobile phones on data provided by PG&E (Docket# 30). UECs for
each of the top five applications in this product class are shown for both manufacturer and test &
teardown engineering inputs in Table 7.22 and Table 7.23, respectively.
Table 7.22 Battery Charger Product Class 2 based on Manufacturer-Supplied Data
Top Applications by
Shipments
CSL0
CSL1
CSL2
CSL3
Shipments
(Units)
(kWh/year)
(kWh/year) (kWh/year) (kWh/year)
Mobile Phones
105,120,000
10.8
5.8
0.7
Answering Machines
20,175,000
13.6
9.6
2.5
Cordless Phones
19,151,000
13.6
9.6
2.5
Portable Video
13,777,000
6.6
4.6
1.2
Game Systems
Digital Cameras
10,879,000
1.4
0.8
0.1
Other Applications
37,119,000
Weighted Average
10.0
6.0
1.2
Notes: Assumes a battery voltage of 3.6 V and battery energy of 3.0 Wh.

Table 7.23 Battery Charger Product Class 2 based on Test & Teardown Data
Top Applications by
Shipments
CSL0
CSL1
CSL2
CSL3
Shipments
(Units)
(kWh/year)
(kWh/year) (kWh/year) (kWh/year)
Mobile Phones
105,120,000
7.7
1.2
Answering Machines
20,175,000
10.7
1.8
Cordless Phones
19,151,000
10.7
1.8
Portable Video
13,777,000
4.4
0.8
Game Systems
Digital Cameras
10,879,000
1.0
0.2
Other Applications
37,119,000
Weighted Average
7.4
1.2
Notes: Assumes a battery voltage of 3.6 V and battery energy of 3.0 Wh.

7.2.4.3 Battery Charger Product Class 3: 4-10 V Battery, <100 Wh


Product class 3 includes a range of products including portable DVD players, toy ride-on
vehicles, and power tools. UECs for each of the top five applications in this product class are
shown in Table 7.24.
Table 7.24 Battery Charger Product Class 3 based on Manufacturer-Supplied Data
Top Applications by
Shipments
CSL0
CSL1
CSL2
CSL3
Shipments
(Units)
(kWh/year)
(kWh/year) (kWh/year) (kWh/year)
Portable DVD
7,140,000
6.4
3.6
0.4
Players
Camcorders
4,206,000
0.8
0.4
0.1
Toy Ride-On
3,548,000
11.3
5.5
2.0
7-27

Vehicles
RC Toys
DIY Power Tools
Other Applications
Weighted Average

2,100,000
1,753,000
4,369,000
-

2.3
10.7
-

1.1
5.4
-

10.0

0.4
2.1
-

5.4

1.0

Notes: Assumes a battery voltage of 7.2 V and battery energy of 10 Wh.

7.2.4.4 Battery Charger Product Class 4: >10 V Battery, <100 Wh


Notebook computers make up more than half of the unit shipments in BC product class 4.
DOE examined two usage profiles for BCs one with significant maintenance time to represent
a desktop replacement user, and one with significant unplugged time to represent a user that
powers their notebook frequently from the battery. UECs for each of the top five applications in
this product class are shown for both manufacturer and test & teardown engineering inputs in
Table 7.25 and Table 7.26, respectively.
Table 7.25 Battery Charger Product Class 4 based on Manufacturer-Supplied Data
Top Applications by
Shipments
CSL0
CSL1
CSL2
CSL3
Shipments
(Units)
(kWh/year)
(kWh/year) (kWh/year) (kWh/year)
Notebooks
40,300,000
32.9
10.0
6.2
Professional Power
11,688,000
79.4
15.9
7.6
Tools
DIY Power Tools
5,259,000
20.5
5.8
3.1
Netbooks
3,700,000
33.1
10.1
6.1
Handheld Vacuums
2,797,000
32.7
9.8
6.5
Other Applications
6,014,000
Weighted Average
39.4
10.5
6.1
Notes: Assumes a battery voltage of 10.8 V and battery energy of 20 Wh.

Table 7.26 Battery Charger Product Class 4 based on Test & Teardown Data
Top Applications by
Shipments
CSL0
CSL1
CSL2
CSL3
Shipments
(Units)
(kWh/year)
(kWh/year) (kWh/year) (kWh/year)
Notebooks
40,300,000
45.5
11.6
5.7
Professional Power
11,688,000
26.3
9.0
9.9
Tools
DIY Power Tools
5,259,000
17.7
5.3
3.4
Netbooks
3,700,000
41.3
10.7
5.8
Handheld Vacuums
2,797,000
52.4
12.9
5.4
Other Applications
6,014,000
Weighted Average
39.4
10.5
6.1
Notes: Assumes a battery voltage of 10.8 V and battery energy of 20 Wh.

7-28

7.2.4.5 Battery Charger Product Class 5: <20 V Battery, 100-3,000 Wh


Toy ride on vehicles make up the majority of shipments in product class 5. UECs for
each of the four applications in this product class are shown for both manufacturer and test &
teardown engineering inputs in Table 7.27 and Table 7.28, respectively.
Table 7.27 Battery Charger Product Class 5 based on Manufacturer-Supplied Data
Shipments
CSL0
CSL1
CSL2
CSL3
Top Applications by
Shipments
(Units)
(kWh/year)
(kWh/year) (kWh/year) (kWh/year)
Toy Ride-On Vehicles
1,774,000
235.4
179.2
127.7
95.9
Marine/Automotive/RV
500,000
103.3
99.1
10.6
7.3
Chargers
Portable O2
50,000
48.0
36.5
26.1
19.6
Concentrators Low
Output
Portable O2
9,000
48.0
36.5
26.1
19.6
Concentrators High
Output
Weighted Average
202.3
158.4
100.0
75.0
Notes: Assumes a battery voltage of 12 V and battery energy of 800 Wh.

Table 7.28 Battery Charger Product Class 5 based on Test & Teardown Data
Shipments
CSL0
CSL1
CSL2
CSL3
Top Applications by
Shipments
(Units)
(kWh/year)
(kWh/year) (kWh/year) (kWh/year)
Toy Ride-On Vehicles
1,774,000
248.4
201.7
Marine/Automotive/RV
500,000
198.2
116.2
Chargers
Portable O2
50,000
50.5
41.1
Concentrators Low
Output
Portable O2
9,000
50.5
41.1
Concentrators High
Output
Weighted Average
232.7
179.3
Notes: Assumes a battery voltage of 12 V and battery energy of 800 Wh.

7.2.4.6 Battery Charger Product Class 6: 20 V Battery, 100-3,000 Wh


Personal transportation equipment dominates product class 6. Most units are charged frequently
and spend a significant amount of time in no-battery mode. DOE assumes users do not unplug
BCs in product class 6. UECs for each of the five applications in this product class are shown in
Table 7.29.

7-29

Table 7.29 Battery Charger Product Class 6 based on Manufacturer-Supplied Data


Top Applications by
Shipments
CSL0
CSL1
CSL2
CSL3
Shipments
(Units)
(kWh/year)
(kWh/year) (kWh/year) (kWh/year)
Electric Scooters
250,000
65.9
39.4
23.5
16.8
Mobility Scooters
192,000
85.3
61.6
36.7
27.4
Lawn Mowers
182,000
45.3
8.8
5.3
2.1
Wheelchairs
166,000
85.3
61.6
36.7
27.4
Motorized Bicycles
150,000
65.9
39.4
23.5
16.8
Weighted Average
69.3
41.9
25.0
18.0
Notes: Assumes a battery voltage of 36 V and battery energy of 384 Wh.

7.2.4.7 Battery Charger Product Class 7: >3,000 Wh


Golf cars have the highest battery energies of any application analyzed by DOE. They
also have the only BCs DOE found with on/off switches. UECs for this application are shown in
Table 7.30.
Table 7.30 Battery Charger Product Class 7 based on Manufacturer-Supplied Data
Top Applications by
Shipments
CSL0
CSL1
CSL2
CSL3
Shipments
(Units)
(kWh/year)
(kWh/year) (kWh/year) (kWh/year)
Golf Cars
214,000
290
250
200
150
Notes: Assumes a battery voltage of 48 V and battery energy of 3,750 Wh.

7.2.4.8 Battery Charger Product Class 8: DC-DC Chargers, <9 V Input


The most common applications that employ USB power (or other 5 V input) are MP3
players and mobile phones. DOE assumes these applications are charged a few times per week
and spend a significant amount of time unplugged. UECs for each of the top five applications in
this product class are shown in Table 7.31.
Table 7.31 Battery Charger Product Class 8 based on Manufacturer-Supplied Data
Top Applications by
Shipments
CSL0
CSL1
CSL2
CSL3
Shipments
(Units)
(kWh/year)
(kWh/year) (kWh/year) (kWh/year)
MP3 Players
39,358,000
1.3
0.8
0.5
Mobile Phones
26,280,000
2.0
1.1
0.7
Digital Cameras
2,720,000
0.3
0.2
0.1
Personal Digital
1,779,000
0.9
0.5
0.3
Assistants
Camcorders
1,402,000
0.1
0.1
0.0
Other Applications
286,000
Weighted Average
1.5
0.9
0.6
Notes: Assumes battery voltage of 3.6 V and battery energy of 2 Wh.

7-30

7.2.4.9 Battery Charger Product Class 9: DC-DC Chargers, 9 V Input


In-vehicle GPSs are by far the most common applications with BCs charged by 12 V
input. DOE assumes these units spend the majority of time unplugged; that is, disconnected from
the cars battery. While the car is running, these units are either in maintenance mode or
charging. UECs for each of the five applications in this product class are shown in Table 7.32.
Table 7.32 Battery Charger Product Class 9 based on Manufacturer-Supplied Data
Top Applications by
Shipments
CSL0
CSL1
CSL2
CSL3
Shipments
(Units)
(kWh/year)
(kWh/year) (kWh/year) (kWh/year)
In-Vehicle GPS
15,320,000
1.2
0.8
0.5
Medical Nebulizers
90,000
9.3
5.7
3.4
Portable O2
50,000
1.2
0.8
0.5
Concentrators Low
Output
Flashlights and
50,000
2.9
1.4
0.7
Lanterns
Portable O2
9,000
1.2
0.8
0.5
Concentrators
High Output
Weighted Average
1.3
0.8
0.5
Notes: Assumes a battery voltage of 3.6 V and battery energy of 5 Wh.

7.2.4.10

Battery Charger Product Class 10: AC-AC, AC Output from Battery

Uninterruptible power supplies are the only application DOE identified in product class
10. They are almost always in maintenance mode. UECs for this application are shown for both
manufacturer and test & teardown engineering inputs in Table 7.33 and Table 7.34, respectively.
Table 7.33 Battery Charger Product Class 10 based on Manufacturer-Supplied Data
Top Applications
Shipments
CSL0
CSL1
CSL2
CSL3
by Shipments
(Units)
(kWh/year)
(kWh/year) (kWh/year) (kWh/year)
Uninterruptible
6,900,000
19.6
6.4
4.0
1.5
Power Supplies
Weighted Average
19.6
6.4
4.0
1.5
Notes: Assumes a battery voltage of 12 V and battery energy of 70 Wh.

Table 7.34 Battery Charger Product Class 10 based on Test & Teardown Data
Top Applications
Shipments
CSL0
CSL1
CSL2
CSL3
by Shipments
(Units)
(kWh/year)
(kWh/year) (kWh/year) (kWh/year)
Uninterruptible
6,900,000
19.6
6.4
Power Supplies
Weighted Average
19.6
6.4
Notes: Assumes a battery voltage of 12 V and battery energy of 70 Wh.

7-31

References
1

Natural Resources Defense Council. "Laptop Computers: How Much Energy Do They Use and How Much Energy

Can We Save?" August 2003.

2
TIAX, LLC. "Energy Consumption by Consumer Electronics in U.S. Residences", Final Report to the Consumer

Electronics Association, January 2007.

3
Nielsen Wireless and Interactive Services. "The State of the Console - Video Game Console Usage - Fourth Quarter

2006". 2007.

7-32

CHAPTER 8. LIFE-CYCLE COST AND PAYBACK PERIOD ANALYSES

TABLE OF CONTENTS
8.1
8.1.1
8.1.2
8.2
8.2.1
8.2.2
8.2.3
8.2.4
8.2.5
8.2.6
8.2.7
8.2.8
8.2.9

8.2.10
8.2.11
8.3
8.3.1
8.3.2
8.3.3
8.4
8.4.1
8.4.2

8.4.3

INTRODUCTION ............................................................................................... 8-1

General Approach for Life-Cycle Cost and Payback Period Analyses ............... 8-1

Overview of LCC and PBP Inputs....................................................................... 8-4

LIFE-CYCLE COST INPUTS ............................................................................ 8-6

Definition ............................................................................................................. 8-6

Total Installed Cost Inputs ................................................................................... 8-7

Operating Cost Inputs .......................................................................................... 8-8

Unit Energy Consumption ................................................................................... 8-9

Electricity Prices .................................................................................................. 8-9

Electricity Price Trend ....................................................................................... 8-10

Gasoline Prices and Trends................................................................................ 8-12

Lifetime.............................................................................................................. 8-13

Discount Rate..................................................................................................... 8-14

8.2.9.1 Residential Discount Rate...................................................................... 8-14

8.2.9.2 Commercial Discount Rate .................................................................... 8-18

Product Energy Efficiency in the Base Case ..................................................... 8-21

Compliance Date of Standard ............................................................................ 8-22

PAYBACK PERIOD INPUTS .......................................................................... 8-23

Definition ........................................................................................................... 8-23

Inputs.................................................................................................................. 8-23

Rebuttable Presumption Payback Period ........................................................... 8-24

LIFE-CYCLE COST AND PAYBACK PERIOD RESULTS .......................... 8-24

Non-Class A External Power Supply Results.................................................... 8-25

Class A External Power Supply Results ............................................................ 8-25

8.4.2.1 Base Case LCC Distributions (Manufacturer Data) .............................. 8-26

8.4.2.2 2.5W Regular AC/DC External Power Supplies (Manufacturer Data) . 8-28

8.4.2.3 18W Regular AC/DC External Power Supplies (Manufacturer Data) .. 8-30

8.4.2.4 60W Regular AC/DC External Power Supplies (Manufacturer Data) .. 8-32

8.4.2.5 120W Regular AC/DC External Power Supplies (Manufacturer Data) 8-34

Battery Charger Results (Representative Units) ................................................ 8-36

8.4.3.1 Base Case LCC Distributions (Manufacturer Data) .............................. 8-37

8.4.3.2 Base Case LCC Distributions (Test/Teardown Data) ............................ 8-40

8.4.3.3 Low Energy, Inductive Battery Chargers (Manufacturer Data) ............ 8-43

8.4.3.4 Low Energy, Inductive Battery Chargers (Test/Teardown Data).......... 8-45

8.4.3.5 Low Energy, Low Voltage Battery Chargers (Manufacturer Data) ...... 8-45

8.4.3.6 Low Energy, Low Voltage Battery Chargers (Test/Teardown Data) .... 8-47

8.4.3.7 Low Energy, High Voltage Battery Chargers (Manufacturer Data)...... 8-49

8.4.3.8 Low Energy, High Voltage Battery Chargers (Test/Teardown Data) ... 8-51

8.4.3.9 Medium Energy, Low Voltage Battery Chargers (Manufacturer Data) 8-53

8.4.3.10
Medium Energy, Low Voltage Battery Chargers (Test/Teardown

Data) 8-55

8-i

8.4.3.11
High Energy Battery Chargers (Manufacturer Data)................. 8-57

8.4.3.12
High Energy Battery Chargers (Test/Teardown Data) .............. 8-59

8.4.3.13
Low Energy, AC Out Battery Chargers (Manufacturer Data) ... 8-59

8.4.3.14
Low Energy, AC Out Battery Chargers (Test/Teardown Data) 8-61

8.4.4 Battery Charger Results (Scaled Product Classes) ............................................ 8-63

8.4.4.1 Base Case LCC Distributions (Manufacturer Data) .............................. 8-64

8.4.4.2 Summarized Life-Cycle Cost and Payback Period Results ................... 8-66

8.5
DETAILED RESULTS AND SENSITIVITY RUNS....................................... 8-67

LIST OF TABLES
Table 8.1.1 Representative Units and Product Classes Analyzed in the LCC and PBP

Analyses........................................................................................................................... 8-3

Table 8.1.2 Scaled Battery Charger Product Classes Analyzed in the LCC and PBP

Analyses........................................................................................................................... 8-3

Table 8.1.3 Summary Information of Inputs for the Life-Cycle Cost and Payback

Period Analyses ............................................................................................................... 8-6

Table 8.2.1 Inputs for Operating Costs........................................................................... 8-8

Table 8.2.2 Electricity Prices by Census Division, 2009.............................................. 8-10

Table 8.2.3 Average Percentage Shares of Household Debt and Equity Types ........... 8-15

Table 8.2.4 Average Nominal and Real Interest Rates for Household Debt Classes ... 8-16

Table 8.2.5 Average Nominal and Real Interest Rates for Household Equity Types .. 8-17

Table 8.2.6 Shares and Interest or Return Rates Used for Household Debt and Equity

Types.............................................................................................................................. 8-17

Table 8.2.7 Variables Used to Estimate Company Discount Rates .............................. 8-19

Table 8.2.8 Real Discount Rates by BC and EPS Ownership Category....................... 8-20

Table 8.2.9 Estimated Share of BC and EPS Purchases by Ownership Category in

Commercial Sector......................................................................................................... 8-21

Table 8.2.10 Average Real Discount Rate by Sector.................................................... 8-21

Table 8.2.11 Base Case Energy Efficiency Market Shares for Class A External Power

Supplies in 2013............................................................................................................. 8-22

Table 8.2.12 Base Case Energy Efficiency Market Shares for Battery Chargers in 2013 8
22

Table 8.2.13 Base Case Energy Efficiency Market Shares for Scaled Battery Charger

Product Classes in 2013 ................................................................................................. 8-22

Table 8.4.1 LCC and PBP Results for 2.5W Regular AC/DC External Power Supplies

(Manufacturer Data)....................................................................................................... 8-30

Table 8.4.2 LCC and PBP Results for 18W Regular AC/DC External Power Supplies

(Manufacturer Data)....................................................................................................... 8-32

Table 8.4.3 LCC and PBP Results for 60W Regular AC/DC External Power Supplies

(Manufacturer Data)....................................................................................................... 8-34

Table 8.4.4 LCC and PBP Results for 120W Regular AC/DC External Power Supplies

(Manufacturer Data)....................................................................................................... 8-36

Table 8.4.5 LCC and PBP Results for Low Energy, Inductive Battery Chargers

(Manufacturer Data)....................................................................................................... 8-45

8-ii

Table 8.4.6 LCC and PBP Results for Low Energy, Low Voltage Battery Chargers

(Manufacturer Data)....................................................................................................... 8-47

Table 8.4.7 LCC and PBP Results for Low Energy, Low Voltage Battery Chargers

(Test/Teardown Data) .................................................................................................... 8-49

Table 8.4.8 LCC and PBP Results for Low Energy, High Voltage Battery Chargers

(Manufacturer Data)....................................................................................................... 8-51

Table 8.4.9 LCC and PBP Results for Low Energy, High Voltage Battery Chargers

(Test/Teardown)............................................................................................................. 8-53

Table 8.4.10 LCC and PBP Results for Medium Energy, Low Voltage Battery Chargers

(Manufacturer Data)....................................................................................................... 8-55

Table 8.4.11 LCC and PBP Results for Medium Energy, Low Voltage Battery Chargers

(Test/Teardown Data) .................................................................................................... 8-57

Table 8.4.12 LCC and PBP Results for High Energy Battery Chargers (Manufacturer

Data)............................................................................................................................... 8-59

Table 8.4.13 LCC and PBP Results for Low Energy, AC Out Battery Chargers

(Manufacturer Data)....................................................................................................... 8-61

Table 8.4.14 LCC and PBP Results for Low Energy, AC Out Battery Chargers

(Test/Teardown Data) .................................................................................................... 8-63

Table 8.4.15 LCC and PBP Results for Low Energy, Medium Voltage Battery Chargers

(Manufacturer Data)....................................................................................................... 8-66

Table 8.4.16 LCC and PBP Results for Medium Energy, High Voltage Battery Chargers

(Manufacturer Data)....................................................................................................... 8-67

Table 8.4.17 LCC and PBP Results from DC-DC, <9V Input Battery Chargers

(Manufacturer Data)....................................................................................................... 8-67

Table 8.4.18 LCC and PBP Results from DC-DC, 9V Input Battery Chargers

(Manufacturer Data)....................................................................................................... 8-67

LIST OF FIGURES

Figure 8.1 Flow Diagram of Inputs for the Determination of LCC and PBP ................. 8-5

Figure 8.2 Residential Sector Electricity Price Trend .................................................. 8-11

Figure 8.3 Commercial Sector Electricity Price Trend................................................. 8-12

Figure 8.4 Gasoline Prices for Years 2009 to 2035 ...................................................... 8-13

Figure 8.5 2.5W Regular AC/DC External Power Supplies: Base Case LCC Distribution

(Manufacturer Data)....................................................................................................... 8-26

Figure 8.6 18W Regular AC/DC External Power Supplies: Base Case LCC Distribution

(Manufacturer Data)....................................................................................................... 8-27

Figure 8.7 60W Regular AC/DC External Power Supplies: Base Case LCC Distribution

(Manufacturer Data)....................................................................................................... 8-27

Figure 8.8 120W Regular AC/DC External Power Supplies: Base Case LCC Distribution

(Manufacturer Data)....................................................................................................... 8-28

Figure 8.9 2.5W Regular AC/DC External Power Supplies: Distribution of Life-Cycle

Cost Impacts (Manufacturer Data)................................................................................. 8-29

Figure 8.10 2.5W Regular AC/DC External Power Supplies: Distribution of PBPs

(Manufacturer Data)....................................................................................................... 8-29

8-iii

Figure 8.11 18W Regular AC/DC External Power Supplies: Distribution of Life-Cycle

Cost Impacts (Manufacturer Data)................................................................................. 8-31

Figure 8.12 18W Regular AC/DC External Power Supplies: Distribution of PBPs

(Manufacturer Data)....................................................................................................... 8-31

Figure 8.13 60W Regular AC/DC External Power Supplies: Distribution of Life-Cycle

Cost Impacts (Manufacturer Data)................................................................................. 8-33

Figure 8.14 60W Regular AC/DC External Power Supplies: Distribution of PBPs

(Manufacturer Data)....................................................................................................... 8-33

Figure 8.15 120W Regular AC/DC External Power Supplies: Distribution of Life-Cycle

Cost Impacts (Manufacturer Data)................................................................................. 8-35

Figure 8.16 120W Regular AC/DC External Power Supplies: Distribution of PBPs

(Manufacturer Data)....................................................................................................... 8-35

Figure 8.17 Low Energy, Inductive Battery Charger: Base Case LCC Distribution

(Manufacturer Data)....................................................................................................... 8-37

Figure 8.18 Low Energy, Low Voltage Battery Charger: Base Case LCC Distribution

(Manufacturer Data)....................................................................................................... 8-38

Figure 8.19 Low Energy, High Voltage Battery Charger: Base Case LCC Distribution

(Manufacturer Data)....................................................................................................... 8-38

Figure 8.20 Medium Energy, Low Voltage Battery Charger: Base Case LCC Distribution

(Manufacturer Data)....................................................................................................... 8-39

Figure 8.21 High Energy Battery Charger: Base Case LCC Distribution (Manufacturer

Data)............................................................................................................................... 8-39

Figure 8.22 Low Energy, AC Out Battery Charger: Base Case LCC Distribution

(Manufacturer Data)....................................................................................................... 8-40

Figure 8.23 Low Energy, Low Voltage Battery Charger: Base Case LCC Distribution

(Test/Teardown Data) .................................................................................................... 8-41

Figure 8.24 Low Energy, High Voltage Battery Charger: Base Case LCC Distribution

(Test/Teardown Data) .................................................................................................... 8-41

Figure 8.25 Medium Energy, Low Voltage Battery Charger: Base Case LCC Distribution

(Test/Teardown Data) .................................................................................................... 8-42

Figure 8.26 Low Energy, AC Out Battery Charger: Base Case LCC Distribution

(Test/Teardown Data) .................................................................................................... 8-42

Figure 8.27 Low Energy, Inductive Battery Chargers: Distribution of Life-Cycle Cost

Impacts (Manufacturer Data)......................................................................................... 8-43

Figure 8.28 Low Energy, Inductive Battery Chargers: Distribution of PBPs

(Manufacturer Data)....................................................................................................... 8-44

Figure 8.29 Low Energy, Low Voltage Battery Chargers: Distribution of Life-Cycle Cost

Impacts (Manufacturer Data) ......................................................................................... 8-46

Figure 8.30 Low Energy, Low Voltage Battery Chargers: Distribution of PBPs

(Manufacturer Data)....................................................................................................... 8-46

Figure 8.31 Low Energy, Low Voltage Battery Chargers: Distribution of Life-Cycle Cost

Impacts (Test/Teardown Data) ...................................................................................... 8-48

Figure 8.32 Low Energy, Low Voltage Battery Chargers: Distribution of PBPs

(Test/Teardown Data) .................................................................................................... 8-48

Figure 8.33 Low Energy, High Voltage BC: Distribution of Life-Cycle Cost Impacts

(Manufacturer Data)....................................................................................................... 8-50

8-iv

Figure 8.34 Low Energy, High Voltage BC: Distribution of PBPs (Manufacturer Data) 8
50

Figure 8.35 Low Energy, High Voltage BC: Distribution of Life-Cycle Cost Impacts

(Test/Teardown Data) .................................................................................................... 8-52

Figure 8.36 Low Energy, High Voltage BC: Distribution of PBPs (Test/Teardown Data)

........................................................................................................................................ 8-52

Figure 8.37 Medium Energy, Low Voltage Battery Chargers: Distribution of Life-Cycle

Cost Impacts (Manufacturer Data)................................................................................. 8-54

Figure 8.38 Medium Energy, Low Voltage Battery Chargers: Distribution of PBPs

(Manufacturer Data)....................................................................................................... 8-54

Figure 8.39 Medium Energy, Low Voltage Battery Chargers: Distribution of Life-Cycle

Cost Impacts (Test/Teardown Data) .............................................................................. 8-56

Figure 8.40 Medium Energy, Low Voltage Battery Chargers: Distribution of PBPs

(Test/Teardown Data) .................................................................................................... 8-56

Figure 8.41 High Energy Battery Chargers: Distribution of Life-Cycle Cost Impacts

(Manufacturer Data)....................................................................................................... 8-58

Figure 8.42 High Energy Battery Chargers: Distribution of PBPs (Manufacturer Data). 8
58

Figure 8.43 Low Energy, AC Out Battery Chargers: Distribution of Life-Cycle Cost

Impacts (Manufacturer Data)......................................................................................... 8-60

Figure 8.44 Low Energy, AC Out Battery Chargers: Distribution of PBPs (Manufacturer

Data)............................................................................................................................... 8-60

Figure 8.45 Low Energy, AC Out Battery Chargers: Distribution of Life-Cycle Cost

Impacts (Test/Teardown Data) ...................................................................................... 8-62

Figure 8.46 Low Energy, AC Out Battery Chargers: Distribution of PBPs

(Test/Teardown Data) .................................................................................................... 8-62

Figure 8.47 Low Energy, Medium Voltage Battery Charger: Base Case LCC Distribution

(Manufacturer Data)....................................................................................................... 8-64

Figure 8.48 Medium Energy, High Voltage Battery Charger: Base Case LCC

Distribution (Manufacturer Data) .................................................................................. 8-65

Figure 8.49 DC-DC, <9V Input Battery Charger: Base Case LCC Distribution

(Manufacturer Data)....................................................................................................... 8-65

Figure 8.50 DC-DC, 9V Input Battery Charger: Base Case LCC Distribution

(Manufacturer Data)....................................................................................................... 8-66

8-v

CHAPTER 8. LIFE-CYCLE COST AND PAYBACK PERIOD ANALYSES


8.1

INTRODUCTION

This chapter describes the analysis the U.S. Department of Energy (DOE) conducts to
evaluate the economic impacts on individual consumers of possible energy conservation
standards for battery chargers (BC) and external power supplies (EPS). New standards usually
decrease operating costs and increase purchase costs for consumers. This chapter describes the
three metrics used in this analysis to determine the impact of standards on individual consumers:
Life-cycle cost (LCC) is the total (discounted) consumer cost over the analysis period
including purchase price, operating costs (including energy expenditures), and
installation costs.
Payback period (PBP) is the number of years it takes a customer to recover the generally
higher purchase price of a more energy-efficient product through the operating cost
savings of using the more energy-efficient product. The PBP is calculated as the change
in first cost divided by the change in operating costs in the first year.
Rebuttable payback period is a special case in which the PBP is calculated based on
laboratory conditions, specifically DOE test procedure inputs. DOE calculates the
aforementioned LCC and PBP using a range of inputs, which are designed to reflect
actual conditions.
Sections 8.2 and 8.3 discuss inputs to the LCC and PBP, respectively. Section 8.3.3
presents the rebuttable presumption PBP and section 8.4 presents the results for the LCC and
PBP. Key variables and calculations are presented for each metric. DOE performs the
calculations discussed here using a series of Microsoft Excel spreadsheets developed for this
rulemaking. Stakeholders are invited to download and examine the spreadsheets, which are
available at:
http://www1.eere.energy.gov/buildings/appliance_standards/residential/battery_external.html.
Appendix 8A presents details and instructions for using the spreadsheets. Appendix 8B
presents sensitivity results and results using different input scenarios. Appendix 8C presents
DOEs application sampling methodology used in the LCC, and Appendix 8D presents further
detail on the calculation of residential discount rates.

8.1.1

General Approach for Life-Cycle Cost and Payback Period Analyses

Recognizing that several inputs to the LCC and PBP analysis are either variable or
uncertain, DOE incorporates Monte Carlo simulation and probability distributions into its LCC
and PBP model. DOE incorporates both Monte Carlo simulation and probability distributions by
using Microsoft Excel spreadsheets combined with Crystal Ball, a commercially available addin program.

8-1

The relationship between increasing selling price and increasing efficiency is the
predominant influence on the LCC and PBP results. However, other factors related to the
characteristics of the consumer using the products also affect the results. Based on the
geographic region, sector, and application in which a consumer uses the BC or EPS, factors such
as energy prices, sales tax, and energy usage can vary. By using the Monte Carlo simulation and
separate sensitivity runs, DOE accounts for this variability. Since many BCs and EPSs are
portable, it is possible for consumers to utilize the same product in both a residential and
commercial setting. While DOE calculates an LCC for both the residential and commercial
sectors, it does not attempt to quantify the extent to which residential consumers use their
products in a commercial setting, or vice versa.
For the LCC and PBP analyses, DOE considers variability in the discount rate. DOE also
models variability in the electricity price by sector, base case efficiency distribution, product
lifetime, sales tax rate, and energy consumption amount by representative unit. By developing
samples, DOE can perform the LCC and PBP calculations and account for the variability in these
inputs among a variety of consumer and regional data. DOE uses the Energy Information
Administrations (EIA) Form 861 from 2008 to develop regional electricity price samples. The
LCC and PBP spreadsheets present the results of the analysis as average values, relative to the
baseline conditions.
The time period used for the LCC and PBP analyses in this rulemaking is the useful life
of the application that the BC or EPS operates. This is because BCs and EPSs are often made
specifically for use with particular consumer products, so their lifetimes relate directly to the
lifetimes of those products. DOE assumes that once the consumer product has reached the end
of its useful life, the user typically discards the associated BC or EPS. Therefore, for each
representative unit, DOE has gathered lifetime values for consumer product applications. DOE
then samples an application based on market-weighting for each representative unit and uses the
mean lifetime associated with that application. In the event that an application lifetime and the
associated EPS or BC lifetime do not coincide, DOE will use the EPS or BC lifetime in its
analysis.
DOE is conducting the LCC and PBP analyses on the baseline BCs and EPSs from the
representative units and representative product classes (representative units) identified in
Chapter 3. Table 8.1.1 shows the set of 10 representative units that DOE is evaluating in this
analysis.

8-2

Table 8.1.1 Representative Units and Product Classes Analyzed in the LCC and PBP
Analyses
Class A External Power Supply Representative Units
2.5 Watt Regular AC/DC
18 Watt Regular AC/DC
60 Watt Regular AC/DC
120 Watt Regular AC/DC
Battery Charger Representative Product Classes
Low Energy, Inductive
Low Energy, Low Voltage
Low Energy, High Voltage
Medium Energy, Low Voltage
High Energy
Low Energy, AC Out

DOE is also conducting the LCC and PBP analyses on the scaled BC product classes.
DOE is considering the scaled BC product classes because these units represent a portion of the
market that may not be captured by the representative units. The scaled EPS product classes,
however, are not analyzed because the representative units chosen span the spectrum of possible
output powers under consideration. More explanation is provided in chapter 5. Table 8.1.2
shows the set of four scaled product classes that DOE is evaluating in this analysis.
Table 8.1.2 Scaled Battery Charger Product Classes Analyzed in the LCC and PBP
Analyses
Battery Charger Scaled Product Classes
Low Energy, Medium Voltage
Medium Energy, High Voltage
DC-DC, <9V Input
DC-DC, 9V Input

There are a number of end-use applications (applications) that use EPSs and BCs from
each representative unit and scaled BC product class outlined in Table 8.1.1 and Table 8.1.2, and
oftentimes the applications are very different. This is because many of the same EPS or BC units
can be used for a variety of applications. Since many of the inputs to the LCC model are
dependent on the particular application, such as product lifetime, DOE considers an array of
popular applications when evaluating each representative unit and scaled BC product class.
Further detail on these applications and DOEs methodology for selecting inputs can be found in
appendix 8C of this preliminary analysis.

8-3

8.1.2

Overview of LCC and PBP Inputs

As mentioned earlier, the LCC represents the total consumer expense over the analysis
period, including purchase expenses, operating costs (including energy expenditures), and
installation costs. DOE discounts future operating costs to the time of purchase and sums them
over the analysis period. There is no residual value for a BC or EPS, since it is often discarded
along with the consumer product that it operates. The PBP represents the number of years it
takes customers to recover the purchase price of more energy-efficient equipment through lower
operating costs. The PBP is calculated as the change in first cost divided by the change in
operating costs in the first year.
DOE categorizes inputs to the LCC and PBP analysis as follows: (1) inputs for
establishing the purchase expense, otherwise known as the total installed cost; and (2) inputs for
calculating the expenses incurred during the operation of the BC or EPS, otherwise known as the
operating cost.
The primary inputs for establishing the life-cycle cost and payback period are:
Manufacturer Selling Price (MSP): As discussed in Chapter 5, the MSP is the final price that
manufacturers sell a product for, including the total cost to produce the product and any
markups the manufacturer applies. The MSP does not account for any distribution or
retail markups, taxes or installation.
Markups: DOE then applies a series of markups to the MSP to convert it to a price that
would be paid by the actual end-use consumer.
Sales Tax: DOE then applies sales tax to convert the end user product price to a final product
price including sales tax. Chapter 6 describes the sales tax markup in detail.
Installation Cost: DOE considers installation costs to be zero for BCs and EPSs because
installation would typically entail a consumer simply unpacking the BC or EPS from the
box it was sold in and connecting the device to mains power and its associated product or
battery. Because the cost of this installation (which may be considered temporary, as
intermittently used devices might be unplugged for storage) is not quantifiable in dollar
terms, DOE considers the installation cost to be zero.
Disposal Cost: DOE considers disposal cost to be zero for BCs and EPSs.
The primary inputs for calculating the operating cost include the following:
Unit Energy Consumption (in kWh/year): The annual site energy consumed by the BC or
EPS at each efficiency level. See Chapter 7 for details of how DOE determines the unit
energy consumption (UEC).
Electricity Prices: DOE uses the average price per kilowatt-hour (i.e., $/kWh) paid by
customers. DOE determines electricity prices using national average residential and
commercial electricity prices for the sample calculation. For the Monte Carlo
distribution, DOE uses average residential and commercial values for 13 regions and
large states. DOE develops all electricity price inputs using 2008 EIA data.

8-4

Electricity Price Trends: DOE uses the EIAs Annual Energy Outlook 2010 Early Release
(AEO2010)i and projections from the Lieberman-Warner Climate Security Act of 2007ii
to forecast electricity prices. For the results presented in this chapter, DOE uses the
AEO2010 reference case to forecast future electricity prices.
Start Year: The year in which the BC or EPS and its associated product are purchased. For
the LCC and PBP analysis, DOE uses 2013 as the start year for all products.
Lifetime: The total years in operation after which the consumer retires the BC or EPS from
service, along with the product it operates.
Discount Rate: The rate at which DOE discounts future expenditures to establish their value
in the year of purchase (2013).
Figure 8.1 depicts the relationships between the installed cost and the operating cost
inputs for the calculation of the LCC and PBP. In this figure, the yellow boxes indicate the
inputs, the green boxes indicate intermediate calculated values, and the blue boxes indicate the
analysis outputs (the LCC and PBP).
Manufacturer
Selling Price
Markups

Final Product
Price

Sales Tax
Payback
Period
Unit Energy
Consumption
Annual
Energy Cost

Lifetime
Operating
Expense

Lifetime

Electricity
Price Trend

Discount
Rate

Start Year

Life-Cycle
Cost

Electricity Prices

Figure 8.1 Flow Diagram of Inputs for the Determination of LCC and PBP
Table 8.1.3 summarizes the input values that DOE uses to calculate the LCC and PBP for
BCs and EPSs. Each row summarizes the total installed cost inputs and operating costs,
including the lifetime, discount rate, and electricity price trend. DOE characterizes several of the
inputs with probability distributions that capture the inputs uncertainty and/or variability in the
Monte Carlo analysis. Table 8.1.3 also lists the chapter of the Technical Support Document that
details the inputs.

8-5

Table 8.1.3
Analyses

Summary Information of Inputs for the Life-Cycle Cost and Payback Period

Factor

Weighted-Average Value

TSD Reference
Section

Manufacturer Selling Price

Total Installed Cost Primary Inputs


Varies with BC and EPS

Chapter 5

Markups
Sales Tax

Varies with BC and EPS


Varies by census region

Chapter 6
Chapter 6

Unit Energy Consumption


Electricity Prices
Electricity Price Trends
Discount Rate

Operating Cost Primary Inputs


Varies with BC and EPS by efficiency level
Vary by sector and census region
Vary with price forecast scenario
Varies with sector

Chapter 7
Chapter 8
Chapter 8
Chapter 8

Lifetime

Varies with BC and EPS

Chapter 3

Sections 8.2 and 8.3 discuss the installed cost and operating cost inputs depicted in Table
8.1.3.
8.2

LIFE-CYCLE COST INPUTS

8.2.1

Definition

The LCC is the total customer cost over the life of a product, including total installed
costs and operating costs. Future operating costs are discounted to the analysis start year (2013)
and summed over the analysis period (the lifetime for each respective representative unit or
scaled BC product class). The LCC is defined by the following equation:
N
OCt
LCC IC
t
t 1 (1 r )
Eq. 8.1
where
LCC
IC
N

OC
r
t

=
=
=
=
=
=
=

life-cycle cost ($),

total installed cost ($),

analysis period,

sum over the analysis period, from year 1 to year N,

operating cost ($),

discount rate,

year for which operating cost is being determined,

DOE expresses all the costs in its LCC and PBP analyses in 2009 dollars. There are no
replacement costs, disposal costs, or residual value associated with BCs or EPSs, so they are
absent from the equation above.

8-6

8.2.2

Total Installed Cost Inputs


The total installed cost to the customer is defined by the following equation:

IC

FPP INST
Eq. 8.2

where
IC

total installed cost, expressed in dollars,

FPP

final product price (i.e., customer price for the product only, including
sales tax), expressed in dollars, and

INST =

installation cost or the customer price to install products, expressed in


dollars. This cost is assumed to be zero for all BCs and EPSs.

In the product price determination (Chapter 6), DOE develops end user product prices
and sales taxes to derive final product prices. Total installed costs are determined using:
End user Product Price ($), and
Sales Tax ($).
The end user product price is the average purchase price a consumer pays before sales tax
for BCs and EPSs. Since consumers often do not purchase BCs or EPSs individually, but rather
as a component of the product they operate, the end user product price for a BC or EPS is
derived as a component of the larger purchase price for the product it is purchased with. The
markups applied to the final product are carried over for the BC or EPS component. The sales
tax represents state and local sales taxes applied to the end user product price. It is a
multiplicative factor that increases the end user product price. DOE calculates the total installed
cost for the BCs and EPSs analyzed based on the following equation:
IC

FPP
PRICE

MU TAX

Eq. 8.3
where
IC
FPP
PRICE
MUTAX

=
=
=
=

total installed cost,

final product price,

end user product price, and

sales tax mark up

Chapter 6 provides detail on the end user product price and sales tax.

8-7

8.2.3

Operating Cost Inputs

The operating cost represents the costs incurred in the operation of the BC or EPS. Table
8.2.1 lists the inputs for operating costs. The lifetime, discount rate, and compliance date of the
standard are required to determine the operating cost and for establishing the operating cost
present value. The electricity consumption for the baseline and other efficiency levels examined
enable comparison of standards operating costs.
Table 8.2.1 Inputs for Operating Costs
Unit Energy Consumption
Electricity Prices
Electricity Price Trends
Discount Rate
Lifetime

The UEC is the estimated energy that a BC or EPS consumes during normal use over the
course of a year. Electricity prices used in the analysis are the price per kilowatt-hour in cents or
dollars (e.g., $/kWh) paid by each customer for electricity. DOE uses electricity price trends to
forecast electricity prices for future year analysis. These trends with the electricity price and
annual UEC are used to calculate the energy cost in each year, which is the operating cost. DOE
defines operating cost by the following equation:
OC

E cons

EP EPT

Eq. 8.4
where
OC
Econs
EP
EPT

=
=
=
=

operating costs,

annual energy consumed,

electricity price, and

electricity price trend factor relative to 2009.

Although operating costs for several other DOE rules have encompassed repair and
maintenance costs, DOE does not consider these costs (which are usually associated with larger
products and appliances) in the LCC for this rulemaking. In making this decision, DOE
recognizes that the service life of a BC or EPS typically exceeds that of the consumer product
with which it is designed to operate. Thus, a consumer would not incur repair or maintenance
costs for a BC or EPS. Also, if a BC or EPS does fail, DOE expects that consumers would
typically discard the EPS and purchase a replacement.
The remainder of this section provides information about each of the above input
variables that DOE used to calculate the operating costs for BCs and EPSs.

8-8

8.2.4 Unit Energy Consumption


BCs and EPSs are unique appliances because they are always used in conjunction with
other products of interest. Most BCs and EPSs are packaged with particular products; thus,
consumers usually do not buy the BC or EPS directly. Instead, for example, consumers obtain
an EPS for a video game system when buying the video game system itself. Thus, although the
LCC and PBP analysis uses the consumer purchase prices of BCs and EPSs, in reality those
prices are a hidden portion of the prices that consumers pay for the products of interest.
Because BCs and EPSs are used in conjunction with other products, their energy
consumption is directly related to the usage of those other products as well as the technologies
that the BC or EPS utilizes. The energy consumption of the analyzed BCs and EPSs is assessed
in further detail in chapter 7, and the technologies of the analyzed BCs and EPSs are assessed in
chapter 3 of the preliminary analysis.
The energy use characterization (Chapter 7) details how DOE determines the UEC for
baseline and standards-compliant products. The UEC varies with the product efficiency. That is,
the energy consumption associated with standards-level products (i.e., products with efficiencies
greater than baseline products) is less than the consumption associated with baseline products.
As such, the UEC decreases with higher improved efficiency. An important input to determining
the energy consumption is the total hours per year that the product is in operation in the different
load states available to the product. With this information, the UEC can be calculated as a
function of its efficiency.
For the LCC and PBP analysis, DOE utilizes different UECs for each efficiency level of
each representative unit and scaled BC product class. Since each representative unit and scaled
BC product class encompasses multiple product applications, each with its own distinct usage
profile and energy consumption levels, DOE uses a shipment-weighted average UEC for each
representative unit and scaled BC product class analyzed in the LCC and PBP analysis for the
sample calculation. For the Monte Carlo simulation analyses, DOE samples an application for
each representative unit and scaled BC product class and uses the UECs associated with that
application. Appendix 8C contains a complete listing of the application inputs associated with
each representative unit and scaled BC product class, including the UECs associated with each
efficiency level for the applications considered and the methodology for sampling these UECs.
8.2.5 Electricity Prices
DOE estimates electricity prices for residential and commercial consumers in each of the
13 regions and large states by using EIA Form 861 data.iii Table 8.2.2 lists the 13 geographic
regions and large states. The EIA Form 861 data are published annually and include annual
electricity sales in kilowatt hours; revenues from electricity sales; and number of consumers for
the residential and commercial sectors for every utility serving final consumers. The calculation
of average electricity prices proceeds in two steps:
1) For each utility, estimate an average residential and commercial price by dividing the
residential or commercial revenues by residential or commercial sales.

8-9

2) Calculate a regional average price, weighting each utility with customers in a region
by the number of residential or commercial consumers served in that region.
The calculation uses the most recent available EIA data at the time the analysis was
conducted, from 2008. Table 8.2.2 shows the results for each geographic region. Because DOE
conducted the LCC and PBP analyses in 2009$, it needed to convert all electricity prices into
2009$. To perform the necessary monetary conversion, DOE uses the consumer price index
(CPI) to convert the electricity prices from 2008$ to 2009$. As described in the following
section on electricity price trends, DOE normalizes energy prices to 2009. Therefore, to forecast
energy prices for any given future year, DOE establishes energy prices for 2009. In Table 8.2.2,
DOE uses data from the AEO2010 Early Release to estimate the electricity prices for 2009.
DOE uses the CPI to adjust for inflation, and then multiplies the electricity price in 2008 by the
ratio of the price in 2009 to that in 2008, as reported in the AEO2010 Early Release, so as to
convert the electricity price to a 2009 estimate.
Table 8.2.2 Electricity Prices by Census Division, 2009
Census Division

Electricity Prices,* 2009


Residential
Commercial
2009$/kWh
2009$/kWh

New England
Middle Atlantic
East North Central
West North Central
South Atlantic
East South Central

$0.170
$0.128
$0.102
$0.086
$0.099
$0.090

$0.151
$0.108
$0.088
$0.071
$0.084
$0.088

West South Central


Mountain
Pacific
New York State
California
Texas
Florida

$0.093
$0.096
$0.102
$0.186
$0.134
$0.127
$0.113

$0.084
$0.082
$0.095
$0.166
$0.120
$0.119
$0.097

U.S. Weighted Average


$0.115
$0.101
* DOE converts dollars to 2009$ by multiplying costs in 2008$ by the ratio of 2009 CPI (214.5) to 2008 CPI
(215.3). DOE converts the price for 2008 to the price in 2009 by multiplying the 2008 price by the ratio of the
average AEO electricity price in 2009 to the average AEO electricity price in 2008.

8.2.6 Electricity Price Trend


The electricity price trend projects the future cost of electricity to 2035. DOE calculates
the LCC and PBP using three separate projections from AEO2010: reference, low economic
growth, and high economic growth. The AEO2010 Early Release was the most recent AEO
available at the time of this preliminary analysis. This AEO2010 contained only the referencecase economic growth projection, however. In order to develop high-economic-growth and low
economic-growth cases for the LCC and PBP analysis, DOE obtained the differences between
the reference case and the high- and low-economic-growth cases in the March 2009 AEO2009,iv

8-10

which includes all three cases. DOE then utilized these differences as scaling factors to generate
high- and low-economic-growth cases based on the AEO2010 Early Release reference case.
DOE also considers an electricity price projection using a carbon cap and trade scenario
based upon the Lieberman-Warner Climate Security Act of 2007. This scenario illustrates an
elevated electricity price projection based on emissions regulations as outlined in the EIAs
S.2191 report accompanying the Act.
These four cases reflect the uncertainty of economic growth in the forecast period. The
high and low growth cases show the projected effects of alternative growth assumptions on
energy markets, while the carbon cap and trade scenario illustrates the possible effects of
emissions regulations. DOE normalizes these four scenarios to the 2009 electricity price, and
then uses that electricity price factor to scale the 2009 electricity prices. Figure 8.2 and Figure
8.3 show the residential and commercial electricity price trends, respectively, based on the three
AEO2010 Early Release projections and the carbon cap and trade scenario projection. The LCC
results presented in this chapter are based on the AEO2010 Early Release reference case.

Price Trend (2009=1)

1.15
1.10
1.05

1.00
Reference

0.95

High Growth

0.90
Start Year: 2013

Low Growth

0.85

Carbon Cap

0.80
2005

2010

2015

2020

2025

2030

2035

Year
Figure 8.2 Residential Sector Electricity Price Trend

8-11

2040

Price Trend (2009=1)

1.15
1.10
1.05
1.00
0.95
0.90
0.85
0.80
0.75
0.70
2005

Reference
High Growth

Start Year: 2013

Low Growth
Carbon Cap

2010

2015

2020

2025

2030

2035

2040

Year

Figure 8.3 Commercial Sector Electricity Price Trend

In the LCC spreadsheet, these electricity price trends are used to project electricity prices
into the future, which are then multiplied by the annual energy usage. The resulting operating
costs are presented in both the LCC spreadsheets and the LCC results tables in this chapter.
8.2.7 Gasoline Prices and Trends
For BCs powered by automotive power sources (scaled BC product class 9), DOE
calculated consumer energy cost savings using gasoline prices rather than electricity prices.
DOE obtained yearly gasoline prices and projections from the AEO2010 Early Release, which
expressed prices per gallon of gasoline. DOE converted these prices into dollars per kilowatthour to project the total gasoline cost for each UEC.
DOE converted the gasoline prices to dollars per kilowatt-hour by using an energy
equivalence conversion factor and an automobile energy conversion efficiency estimate. An
estimate of 33.705 kWh/gallon of gasoline was used to estimate fully-efficient conversion of
energy.v From this estimate, DOE applied an average automobile energy conversion efficiency
of 21-percent, which represents the typical efficiency in converting gasoline to electric power.vi
Using these figures, DOE estimates 7.08kWh/gallon when powering a BC through an automotive
power source. DOE then converted the AEO2010 Early Release gasoline price estimates
(expressed as dollars per gallon) into dollars per kilowatt-hour. The resulting gasoline prices per
kilowatt-hour are expressed in Figure 8.4.

8-12

Gas Price (2009$/kWh)

$0.60
$0.55
$0.50
$0.45
$0.40
Start Year: 2013

$0.35
$0.30
2005

2010

2015

2020

2025

2030

2035

2040

Year
Figure 8.4 Gasoline Prices for Years 2009 to 2035

8.2.8 Lifetime
DOE considers the lifetime of a BC or EPS to be from the moment it is purchased for enduse up until the time when it is permanently retired from service. Because the typical BC or EPS
is purchased for use with a single associated application, DOE assumes that it will remain in
service for as long as the application does. Since there are multiple applications with different
lifetimes for a single representative unit or scaled BC product class, a shipment-weighted
average lifetime is calculated from the relevant applications for that representative unit or scaled
BC product class in the Sample Calculation. The Monte Carlo simulation analysis samples a
lifetime from the relevant applications for each representative unit or scaled BC product class.
Since each application has its own lifetime estimate, this provides a distribution of lifetime
estimates for a given representative unit or scaled BC product class. Further detail on this
application sampling methodology can be found in Appendix 8C to this preliminary analysis.
Chapter 3 of this TSD contains the lifetimes for the selected applications and the methodology
for deriving these estimates.
Even though many of the technology options to improve BC and EPS efficiencies may
result in an increased useful life for the BC or EPS, the lifetime of the BC or EPS is still directly
tied to the lifetime of its associated application. Even if an EPS or BC has a lifetime that exceeds
the lifetime of its application, the typical consumer will not use the EPS or BC once the
application has been discarded. For this reason, the baseline and standard level designs for the
LCC and PBP analyses all use the same lifetime estimate. Further detail on product lifetimes
and how they relate to applications can be found in chapter 3 of this TSD.

8-13

8.2.9 Discount Rate


The discount rate is the rate at which DOE discounts future expenditures to establish their
present values. In the LCC analysis, DOE derives the discount rates separately for residential
and commercial consumers. For residential consumers, DOE estimates the discount rate by
looking across all possible debt or asset classes that might be used to purchase BCs or EPSs. For
the commercial consumers, DOE estimates the cost of capital for commercial companies by
examining both debt and equity capital, and develops an appropriately weighted average of the
cost to the company of equity and debt financing.
8.2.9.1 Residential Discount Rate
DOEs approach for the residential discount rate involves identifying all possible debt or
asset classes that might be used to purchase BCs or EPSs, including household assets that might
be affected indirectly.1 DOE does not include debt from primary mortgages and equity of assets
considered non-liquid (such as retirement accounts), since these would likely not be used to
finance BC or EPS purchases. DOE estimates the average shares of the various debt and equity
classes in the average U.S. household equity and debt portfolios using the Federal Reserves
Survey of Consumer Finances (SCF) data for 1989, 1992, 1995, 1998, 2001, 2004, and 2007.vii
Table 8.2.3 shows the average shares of each considered class. DOE uses the mean share of each
class across the seven survey years (18 years) as the basis for estimating household financing of
BCs and EPSs.

An indirect effect would arise if a household sold assets to pay off a loan or credit card debt that might have been
used to finance the actual EPS or BC purchase.

8-14

Table 8.2.3 Average Percentage Shares of Household Debt and Equity Types
Type

1989 SCF 1992 SCF 1995 SCF 1998 SCF 2001 SCF 2004 SCF 2007 SCF
%
%
%
%
%
%
%

Mean
%

Home Equity Loans

2.2

1.9

0.9

1.0

1.2

2.9

2.5

1.8

Credit Cards

0.0

2.2

2.6

2.2

1.7

2.0

2.5

1.9

Other Installment Loans

3.1

1.8

1.4

1.7

1.1

1.4

1.2

1.7

Other Residential Loans

0.0

7.0

5.3

4.4

3.1

5.9

7.3

4.7

Other Line of Credit

1.2

0.6

0.4

0.0

0.3

0.5

0.3

0.5

Checking Accounts

6.3

4.8

4.9

4.0

3.7

4.3

3.5

4.5

Savings and Money Market

20.9

19.3

14.3

13.1

14.5

15.4

13.3

15.8

Certificate of Deposit (CD)

15.8

12.0

9.5

7.1

5.5

6.0

6.6

8.9

2.4

1.7

2.2

1.1

1.2

0.9

0.7

1.5

Bonds

15.0

12.6

10.7

7.1

8.0

8.6

6.8

9.8

Stocks

24.4

24.7

26.4

37.7

38.1

28.4

29.3

29.8

8.7

11.4

21.3

20.5

21.6

23.7

26.0

19.1

100.0

100.0

100.0

100.0

100.0

100.0

100.0

100.0

Savings Bond

Mutual Funds
Total

DOE estimates interest or return rates associated with each type of equity and debt. The
data source for the interest rates for loans, credit cards, and lines of credit is the Federal Reserve
Boards SCF in 1989, 1992, 1995, 1998, 2001, 2004, and 2007. The top half of Table 8.2.4
shows the average nominal interest rates in each year and the inflation rates used to calculate real
rates (using the Fisher formula).2 For home equity loans, DOE calculates effective interest rates
using a tax adjustment since interest on such loans is tax deductible. The bottom half of the table
shows the average effective real interest rates in each year and the mean rate across all the years.
Since the interest rates for each debt carried by households in these years were established over
18 years, DOE believes they are representative of rates that may be in effect in 2013.

Fisher formula is given by: Real Interest Rate = [(1 + Nominal Interest Rate) / (1 + Inflation Rate)] 1.

8-15

Table 8.2.4 Average Nominal and Real Interest Rates for Household Debt Classes
Type

1989 SCF 1992 SCF 1995 SCF 1998 SCF 2001 SCF 2004 SCF 2007 SCF

Mean

Nominal Interest Rates %


Home Equity Loans

8.4

6.4

7.0

6.8

6.0

3.6

5.8

6.3

14.1

14.0

13.9

11.1

12.4

13.1

Other Installment Loans

8.4

7.5

8.5

7.6

8.3

8.1

8.8

8.2

Other Residential Loans

9.8

8.9

8.3

8.0

7.7

6.0

6.4

7.9

Other Line of Credit

14.4

12.5

11.6

11.6

9.4

7.6

9.6

11.0

Inflation rate

4.82

3.01

2.83

1.56

2.85

2.66

2.85

Credit Cards*

Real Interest Rates %


Home Equity Loans

4.5

4.1

4.8

5.4

3.7

1.5

3.5

3.9

11.0

12.1

10.9

8.4

9.6

10.4

Other Installment Loans

4.6

5.7

6.5

6.4

6.3

6.1

6.8

6.1

Other Residential Loans

4.6

5.6

5.3

6.1

4.7

3.2

3.4

4.7

Other Line of Credit

9.3

9.3

8.4

9.7

6.2

4.9

6.7

7.8

Credit Cards*

*No interest rate data available for credit cards in 1989 or 1992.

To account for variation among new households, DOE samples a rate for each household
from a distribution of rates for each of the above debt classes. DOE develops a probability
distribution of interest rates for each debt class based on the SCF data. Appendix 8D presents
the probability distribution of interest rates for each debt class that DOE used in the LCC and
PBP analyses.
Similar rate data are not available from the SCF for the asset classes, so DOE derives
data for these classes from national historical data. The interest rates associated with certificates
of deposit (CDs),viii savings bonds,ix and bonds (AAA corporate bonds) x are from Federal
Reserve Board time-series data 1979-2009. DOE assumes rates on checking accounts to be zero.
Rates on savings and money market accounts are from Cost of Savings Index data covering
1984-2009.xi The rates for stocks are the annual returns on the Standard and Poors (S&P) 500
1979-2009.xii The mutual fund rates are a weighted average of the stock rates (two-thirds
weight) and the bond rates (one-third weight) in each year from 1979 to 2009. DOE adjusts the
nominal rates to real rates using the annual inflation rate in each year. Average nominal and real
interest rates for the classes of assets are shown in Table 8.2.5. Since the interest and return rates
for each asset type cover a range of time, DOE believes they are representative of rates that may
be in effect in 2013.

8-16

Table 8.2.5 Average Nominal and Real Interest Rates for Household Equity Types
Type
Checking Accounts

Average Nominal Rate %


-

Average Real Rate %


-

Savings and Money Market

5.3

2.3

CDs

6.4

2.4

Savings Bonds

8.3

4.2

Bonds

8.4

4.3

Stocks

12.8

8.6

Mutual Funds

11.3

7.1

To account for variation among new households, DOE samples a rate for each household
from a distribution of rates for each of the above asset types. DOE develops a normal probability
distribution of interest rates for each asset type by using the mean value and standard deviation
from the distribution. Appendix 8D presents the probability distribution of interest rates for each
asset type that DOE used in the LCC and PBP analysis.
Table 8.2.6 summarizes the mean real effective rates of each type of equity or debt. DOE
determines the average share of each debt and asset using SCF data for 1989, 1992, 1995, 1998,
2001, 2004, and 2007. Each year of SCF data provides the debt and asset shares for U.S.
households. DOE averages the debt and asset shares over the seven years of survey data to
arrive at the shares shown below. The average rate across all types of household debt and equity,
weighted by the shares of each class, is 5.6-percent.
Table 8.2.6 Shares and Interest or Return Rates Used for Household Debt and Equity
Types
Average Share of Household
Debt Plus Equity*
%
1.8
1.9
1.7

Mean Effective Real Rate**


%
3.9
10.4
6.1

Other Residential Loans


Other Line of Credit
Checking Accounts
Savings and Money Market Accounts
CDs
Savings Bonds
Bonds

4.7
0.5
4.5
15.8
8.9
1.5
9.8

4.7
7.8
0.0
2.3
2.4
4.2
4.3

Stocks
Mutual Funds

29.8
19.1

8.6
7.1

100.0
Total/Weighted-Average Discount Rate
* Not including primary mortgage or retirement accounts.

** Adjusted for inflation and, for home equity loans, loan interest tax deduction.

5.6

Type
Home Equity Loans
Credit Cards
Other Installment Loans

8-17

8.2.9.2 Commercial Discount Rate


Most companies use both debt and equity capital to fund investments; for most
companies, therefore, the cost of capital is the weighted average of the cost to the firm of equity
and debt financing.xiii
DOE estimates the cost of equity financing using the Capital Asset Pricing Model
(CAPM). The CAPM, among the most widely used models to estimate the cost of equity
financing, assumes that the cost of equity is proportional to the amount of systematic risk
associated with a firm. For example, the cost of equity financing tends to be high when a firm
faces a large degree of systematic risk, and the cost tends to be low when the firm faces a small
degree of systematic risk.
The degree of systematic risk facing a firm and the subsequent cost of equity financing
are determined by several variables, including the risk coefficient of a firm (beta, or B), the
expected return on risk-free assets (Rf), and the additional return expected on assets facing
average market risk (known as the equity risk premium, or ERP). The beta indicates the degree
of risk associated with a given firm, relative to the level of risk (or price variability) in the
overall stock market. Betas usually vary between 0.5 and 2.0. A firm with a beta of 0.5 faces
half the risk of other stocks in the market; a firm with a beta of 2.0 faces twice the overall stock
market risk.
Following this approach, the cost of equity financing for a particular company is by the
equation:
ke R f
ERP
Eq. 8.5
where
ke
Rf

ERP

=
=
=
=

the cost of equity for a company, expressed in dollars,

the expected return of the risk-free asset, expressed in dollars,

the risk coefficient, and

the expected equity risk premium, expressed in dollars.

The cost of debt financing (kd) is the yield or interest rate paid on money borrowed by a
company (raised, for example, by selling bonds). As defined here, the cost of debt includes
compensation for default risk and excludes deductions for taxes.
DOE estimates the cost of debt for companies by adding a risk adjustment factor to the
current yield on long-term corporate bonds (the risk-free rate). This procedure is used to
estimate current and future company costs to obtain debt financing. The adjustment factor is
based on indicators of company risk, such as credit rating or variability of stock returns.

8-18

The discount rate of companies is the weighted average cost of debt and equity financing,
less expected inflation. DOE estimates the discount rate using the equation:

ke

we

kd

wd

Eq. 8.6
where
k
ke and kd

=
=

we and wd

the (nominal) cost of capital,


the expected rates of return on equity and debt,
respectively, and
the proportion of equity and debt financing, respectively.

The real discount rate deducts expected inflation from the nominal rate.
The expected return on risk-free assets, or the risk-free rate, is defined by the current
yield on long-term (20-year) Government bonds, as suggested by Ibbotsons Associatesxiv and
Damodaran.xii The ERP represents the difference between the expected (average) stock market
return and the risk-free rate. As Table 8.2.7 shows, DOE uses an ERP estimate of 5.5-percent,
which it took from the Damodaran Online site (a private website associated with New York
Universitys Stern School of Business, which aggregates information on corporate finance,
investment, and valuation).xv
DOE calculates an expected inflation of 1.6-percent from the average of the projected
change in gross domestic product (GDP) prices in the Economic Report of the President.xvi DOE
obtained the cost of debt, percent debt financing, and systematic firm risk from the Damodaran
Online website. Table 8.2.7 shows average values across all private companies. However, the
cost of debt, percent debt financing, and systematic firm risk vary by sector.
Table 8.2.7 Variables Used to Estimate Company Discount Rates
Variable

Risk-Free Asset Return


Equity Risk Premium
Expected Inflation
Cost of Debt (After Tax)
Percent Debt Financing
Systematic Firm Risk

Symbol

Rf
ERP
R
kd
wd
B

Average
Value
%
5.3
5.5
1.6
6.0
27.9
1.0

Source

Damodaran Online
Damodaran Online
2010 Economic Report of the President
Damodaran Online
Damodaran Online
Damodaran Online

In the commercial building sector, BCs and EPSs are purchased and owned by
commercial companies, industrial companies and the Government. DOE uses a sample of 4,774
companies drawn from these owner categories to represent BC and EPS purchasers. It took the
sample from the list of companies included in the Value Line investment surveyxvii and listed on
the Damodaran Online website. DOE obtained the cost of debt, the firm beta, the percent of debt
and equity financing, the risk-free return, and the equity risk premium from Damodaran Online.

8-19

DOE estimates the cost of debt financing for these companies from the long-term
Government bond rate and the standard deviation of the stock price. For Government-office
type owners, the discount rate represents an average of the Federal rate and the State and local
bond rate. DOE drew the Federal rate directly from the U.S. Office of Management and Budget
discount rate for investments in Government building energy efficiency.xviii DOE estimates the
State and local discount rate from the interest rate on State and local bonds between 1979 and
2009.ix DOE uses this information to estimate the weighted-average cost of capital for the
sample of companies included in the company database.
The cost of capital may be viewed as the discount rate that should be used to reduce the
future value of typical company project cash flows. It is a nominal discount rate, since
anticipated future inflation is included in both stock and bond expected returns. Deducting
expected inflation from the cost of capital provides estimates of the real discount rate by
ownership category (see Table 8.2.8). The mean real discount rate for these companies varies
between 3.1-percent (Government offices) and 8.0-percent (industrial companies).
Table 8.2.8 Real Discount Rates by BC and EPS Ownership Category
Ownership Category

SIC
Codes*

Mean Real
Standard
Number of
Discount Rate
Deviation Observations
%
%
Industrial Companies
14
8.0
1.9
2,726
Commercial Trade
5
7.7
1.8
280
Commercial Finance, Insurance & Real Estate
6
6.6
1.7
852
Commercial Services
78
7.6
1.7
916
Government Offices
N/A
3.1
1.2
N/A
* SIC Codes refer to the U.S. Standard Industrial Classification system.

Source: Navigant Consulting, Inc. calculations based on firms sampled from the Damodaran Online website.

DOEs approach for estimating the cost of capital provides a measure of the discount rate
spread as well as the average discount rate. DOE infers the discount rate spread by ownership
category from the standard deviation, which ranges between 1.2-percent and 1.9-percent (Table
8.2.8).
To estimate the share of each ownership category in total commercial sector purchases of
BCs and EPSs, DOE uses the share of each category in total paid employees. DOE uses the most
current data reported by the U.S. Census Bureau to see the number of employees by industry, xix
for the Federal government, xx and for State and Local governments.xxi DOE uses the number of
employees as a proxy for the share of each ownership category because the prevalence of most
BCs and EPSs will vary depending on the number of individual users for a given device. Table
8.2.9 presents the estimated shares of commercial BC and EPS purchases by ownership category.

8-20

Table 8.2.9 Estimated Share of BC and EPS Purchases by Ownership Category in


Commercial Sector
Ownership Category
Industrial Companies

Percent (%)
23.3

Commercial Trade
Commercial Finance, Insurance & Real Estate
Commercial Services
Government Offices
Source: U.S. Census Bureau.

12.8
6.4
43.6
13.9

DOE estimates discount rate distributions for the different sectors as a weighted average
of the distributions for the different ownership types. Table 8.2.10 summarizes the weighted
average real discount rates in the residential and commercial sectors.
Table 8.2.10 Average Real Discount Rate by Sector
Sector
Residential
Commercial

Discount Rate
%
5.6
7.0

8.2.10 Product Energy Efficiency in the Base Case


For purposes of conducting the LCC analysis, DOE analyzed candidate standard levels
relative to a base case (i.e., a case without new energy conservation standards). This requires an
estimate of the distribution of product efficiencies in the base case (i.e., what consumers would
have purchased in 2013 in the absence of new standards). Rather than analyzing the impacts of a
particular standard level assuming that all consumers will purchase products at the baseline
efficiency level, DOE conducted the analysis by taking into account the breadth of product
energy efficiencies that consumers are expected to purchase under the base case.
As discussed in section 8.1.1, DOEs approach for conducting the LCC analysis for BCs
and EPSs relied on developing samples of consumers that use each of the products, and using a
Monte Carlo simulation technique to perform the LCC calculations on the consumers in the
sample. DOE assigned each consumer in the sample a unique product energy efficiency taken
from the estimated base case distribution of product energy efficiencies in the compliance year.
The energy efficiency distributions used for each application and each representative unit or
scaled BC product class are presented in chapter 3 of the TSD. The applications assigned to each
representative unit and scaled BC product class, and their respective energy efficiency
distribution, is explained in further detail in Appendix 8C.
DOE calculated the efficiency distributions by evaluating tested efficiency data for each
representative unit and scaled BC product class. The tested efficiencies were then binned into
the appropriate standards-level and weighted by the shipments of the application tested. DOE
assumed that the current averages for BCs are reasonable to use as a base case for 2013, but
assumed a modest improvement in EPS efficiency from 2009 to 2013. This is because Europe
8-21

recently passed efficiency standards requiring EPSs to meet the EISA level (CSL 0), and the
efficiency requirement is set to rise to Energy Star 2.0 (CSL 1) by April 2011. DOE assumes
this will impact the American market by shifting EPS efficiencies higher by year 2013. Further
detail on this analysis can be found in chapter 9 of the TSD. Table 8.2.11 and Table 8.2.12 show
the application-weighted efficiency distributions that were used for each representative unit.
Table 8.2.13 shows the application-weighted efficiency distributions that were used for each
scaled BC product class.
Table 8.2.11 Base Case Energy Efficiency Market Shares for Class A External Power
Supplies in 2013
Energy
Efficiency Level
Baseline
1
2
3
4

2.5 Watt
34.0
50.7
12.0
3.3
0.0

Market Share (%)


18 Watt
60 Watt
34.0
34.0
50.7
50.7
12.0
12.0
3.3
3.3
0.0
0.0

120 Watt
34.0
50.7
12.0
3.3
0.0

Table 8.2.12 Base Case Energy Efficiency Market Shares for Battery Chargers in 2013
Energy
Efficiency Level
Baseline
1
2
3

Low E,
Inductive
75.0
12.5
12.5
0.0

Low E,
Low V
39.2
58.0
2.8
0.0

Market Share (%)


Low E,
Medium E,
High V
Low V
18.7
43.4
43.0
16.3
38.3
40.3
0.0
0.0

High E
50.0
20.0
30.0
0.0

Low E,
AC Out
66.7
22.2
11.1
0.0

Table 8.2.13 Base Case Energy Efficiency Market Shares for Scaled Battery Charger
Product Classes in 2013
Energy
Efficiency Level
Baseline
1
2
3

Low E,
Medium V
34.0
54.0
12.0
0.0

Market Share (%)


Medium E,
DC-DC,
High V
<9V Input
50.0
39.2
20.0
58.0
30.0
2.8
0.0
0.0

DC-DC,
9V Input
39.2
58.0
2.8
0.0

8.2.11 Compliance Date of Standard


The compliance date is the date when a new standard becomes operative, i.e., the date by
which BC and EPS manufacturers must manufacture products that comply with the standard.
DOEs publication of a final rule in this standards rulemaking is scheduled for completion in
July 2011. The compliance date for amended EPS standards is July 1, 2013. (42 U.S.C.
6295(u)(3)(D)(i)(II)(bb) The compliance date for BC standards is also targeted to be July 1,
2013. DOE calculates the LCCs for all consumers as if each would purchase a new product in
the year the standard takes effect (2013). However, DOE bases the cost of the equipment on the
most recent available data; all dollar values are expressed in 2009$.

8-22

8.3

PAYBACK PERIOD INPUTS

8.3.1

Definition

The PBP is the amount of time it takes the consumer to recover the assumed higher
purchase cost of a more energy-efficient product as a result of lower operating costs.
Numerically, the PBP is the ratio of the increase in purchase cost (i.e., from a less efficient
design to a more efficient design) to the decrease in annual operating expenditures. This type of
calculation is known as a simple PBP, because it does not take into account changes in
operating cost over time or the time value of money. That is, the calculation is done at an
effective discount rate of zero percent.
The equation for PBP is:

IC
OC

PBP

Eq. 8.7
where
PBP
IC

=
=

OC

payback period in years,


difference in the total installed cost between the more efficient
standard level product (efficiency levels 1, 2, etc.) and the baseline
(efficiency level 0) product, and
difference in annual operating costs.

PBPs are expressed in years. PBPs greater than the life of the product mean that the
increased total installed cost of the more efficient product is not recovered in reduced operating
costs over the lifetime of that product. Because all BC and EPS designs in the LCC and PBP
analyses save energy and thus yield a positive OC, PBPs that are negative or equal to zero
indicate that the total installed cost of the equipment that meets the higher EL is less than that of
the baseline.
8.3.2

Inputs

The data inputs to the PBP calculation are the total installed cost of the product to the
customer for each candidate standard level (CSL) and the annual (first year) operating costs for
each CSL. The only input to the total installed cost is the final product price since the
installation cost is assumed to be zero for BCs and EPSs. The inputs to the operating costs are
the BCs or EPSs UEC and the cost of electricity. The PBP calculation uses the same inputs as
the LCC calculation described in section 8.2, except that electricity price trends are not required.
Since the PBP is a simple (undiscounted) PBP, the required electricity cost is only for the year
in which a new energy conservation standard is to take effectin this case, 2013. The electricity
price DOE uses in the PBP calculation is the price projected for 2013, expressed in 2009$, but
not discounted to 2009. DOE does not use discount rates in the PBP calculation.

8-23

8.3.3 Rebuttable Presumption Payback Period


Section 325(o)(2)(B)(iii) of EPCA establishes a rebuttable presumption that a standard
for BCs or EPSs is economically justified if the Secretary finds that the additional cost to the
consumer of purchasing a product complying with an energy conservation standard level will be
less than three times the value of the energy. . . savings during the first year that the consumer
will receive as a result of the standard, as calculated under the applicable test procedure . . . .
(42 U.S.C. 6295(o)(2)(B)(iii)) This rebuttable presumption test is an alternative path to
establishing economic justification, as compared to consideration of the seven factors set forth in
42 U.S.C. 6295(o)(2)(B)(i)(I)-(VII).
For EPSs and BCs, energy savings calculations in the LCC and PBP analyses use both
the relevant test procedures as well as the relevant usage profiles. Because DOE calculates
payback periods in a methodology consistent with the rebuttable presumption test for EPSs and
BCs in the LCC and payback period analyses, DOE is not performing a stand-alone rebuttable
presumption analysis, as it is already embodied in the LCC and PBP analyses.
8.4

LIFE-CYCLE COST AND PAYBACK PERIOD RESULTS

This section presents LCC results for each BC and EPS representative unit and scaled BC
product class that DOE considers. This section uses the terms positive LCC savings and
negative LCC savings. When a standard results in positive LCC savings, the life cycle cost
of the standards-compliant BC or EPS is less than the life-cycle cost of the baseline BC or EPS
and the consumer will benefit. A consumer is adversely affected when a standard results in
negative LCC savings (i.e., when the life-cycle cost of the standards-compliant BC or EPS is
higher than the life-cycle cost of the baseline BC or EPS). As mentioned previously, DOE
characterized the uncertainty and variability of many of the inputs to the analysis with
probability distributions and then used a Monte Carlo simulation technique to perform the LCC
and PBP calculations. DOE calculated the average LCC and LCC savings and the median and
average PBP for each of the CSLs.
DOE calculated LCC savings and PBPs at each efficiency level relative to the base case
products that it assigned to the consumers. For some consumers, DOE assigned base case
products that are more energy efficient than some of the standard levels. If a consumer was
assigned a product energy efficiency that is greater than or equal to the energy efficiency of the
standard level under consideration, the LCC calculation reveals that this consumer is not
impacted by an increase in product energy efficiency to the standard level, and will experience
LCC savings of $0. For that reason, the average LCC impacts are not equal to the difference
between the LCC of a specific standard level and the LCC of the baseline products. The PBP
calculations, however, only consider users who would be affected by the standard.
This section presents LCC and PBP results for the residential sector using the energy
price forecast in the Reference case from the AEO2010 Early Release. Appendix 8B presents
results for the commercial sector, results using the energy price forecasts in the Low, High, and
Carbon Cap and Trade growth cases, and results for high-savings and low-savings input
scenarios.

8-24

In the subsections below, DOE presents figures showing the distribution of LCCs in the
base case for each representative unit and scaled BC product class. Also presented below are
figures showing the distribution of LCC impacts and the distribution of PBPs. The distributions
of LCCs are presented as frequency charts that show the distribution of LCCs with their
corresponding probability of occurrence. DOE used box-and-whisker plots to show the LCC
savings and PBP impacts at each standard level considered. DOE generated the figures for the
distributions from a Monte Carlo simulation run based on 1,000 samples.
8.4.1 Non-Class A External Power Supply Results
As indicated in chapter 5, DOE did not generate engineering analysis outputs for NonClass A EPSs in the preliminary analysis. The engineering analysis will consider Non-Class A
EPSs in the NOPR phase. At that time, DOE will examine the LCC and PBP of these products.
Please refer to the Determination Concerning the Potential for Energy Conservation Standards
for Non-Class A External Power Supplies published on November 3, 2009 (74 FR 56928) for the
most current LCC and PBP analysis for these products.
8.4.2 Class A External Power Supply Results
DOE conducted a life-cycle cost analysis on four representative units for Class A EPSs at
different levels of output power: 2.5 watts, 18 watts, 60 watts, and 120 watts. As discussed in
chapter 5, DOE created two sets of cost-efficiency curves for each of the four representative
units using data from manufacturer interviews and data from DOEs own efficiency tests and
cost teardowns (tests/teardowns). For the data from manufacturer interviews, DOE aggregated
the results from various manufacturers. As a result, DOE had to normalize the MSP values
across manufacturers. For this reason, the MSPs are expressed as an incremental cost increase
over the baseline level, where the baseline MSP is assigned a value of $0. As a result, the
average installed prices for Class A EPSs are not indicative of the total installed cost at each
CSL, and thus the average LCC results are also not indicative of the total life-cycle cost for each
CSL. However, the LCC savings, PBP calculations, and percentage of consumers with a net cost
or net benefit are unaffected by using these MSPs since they consider the incremental change
from one CSL to the other.
As explained in chapter 5, several of the costs resulting from DOEs teardown analysis
resulted in a negative-sloping cost-efficiency curve. This indicates that a unit with a higher
efficiency would have a lower purchase price than a unit with a lower efficiency. A negativesloping cost-efficiency curve will result in positive LCC savings at every standards-level because
the consumer experiences both a lower installed cost and lower operating costs. Any LCC or
PBP analysis on these units would indicate the highest CSL as having the greatest benefit to the
consumer. Only the 60W EPS had a higher price for the best-in-market design (CSL 3) than for
the baseline design (CSL 0), but this price was lower than the CSL 1 designs price. While the
results for the 60W test/teardown data look similar to the results using the manufacturer data,
DOE only presents the manufacturer data since it does not contain any negative-sloping costefficiency curves. Similarly, DOE does not present LCC or PBP results for any test/teardown
data collected for Class A EPSs, and defers to the data aggregated from manufacturer interviews.

8-25

8.4.2.1 Base Case LCC Distributions (Manufacturer Data)


Figure 8.5 through Figure 8.8 show the frequency charts for the base case LCC for the
four Class A EPS representative units. Since the base case considers the different efficiency
levels at which consumers currently purchase products, the base case LCC distribution is
composed of several standards-levels. If all consumers purchased products at CSL 0 in the base
case, then the LCC distribution would be composed entirely of the LCC at CSL 0. However, the
LCC distribution in the base case shows the LCC results in proportion to how many users
currently purchase products at each standards-level (CSLs 0 through 4).
To find the appropriate distribution of LCCs in the base case, the total LCC was
calculated for each standards-level considered (CSLs 0 through 4). Then these distributions were
weighted by the market shares for each standards-level in the base case, as discussed in section
8.2.10. The combined chart shows the aggregated LCC distribution across all CSLs in the base
case.

8%
7%
6%
5%
4%
3%
2%
1%
0%

CSL 4
CSL 3

CSL 2
CSL 1

CSL 0

0.10
0.30
0.60
0.80
1.00
1.30
1.50
1.70
2.00
2.20
2.50
2.70
2.90
3.20
3.40
3.60
3.90

Percent of Consumers (%)

Total LCC in the Base Case


(weighted by market distribution)

Weighted Total LCC ($)

Figure 8.5 2.5W Regular AC/DC External Power Supplies: Base Case LCC Distribution
(Manufacturer Data)

8-26

9%
8%
7%
6%
5%
4%
3%
2%
1%
0%

CSL 4
CSL 3

CSL 2
CSL 1

CSL 0

0.30
1.30
2.40
3.50
4.60
5.70
6.80
7.80
8.90
10.00
11.10
12.20
13.30
14.30
15.40
16.50
17.60

Percent of Consumers (%)

Total LCC in the Base Case


(weighted by market distribution)

Weighted Total LCC ($)

Figure 8.6 18W Regular AC/DC External Power Supplies: Base Case LCC Distribution
(Manufacturer Data)

8%
7%
6%
5%
4%
3%
2%
1%
0%

CSL 4
CSL 3

CSL 2
CSL 1

CSL 0

2.30
2.80
3.30
3.80
4.30
4.80
5.30
5.80
6.30
6.80
7.30
7.80
8.30
8.80
9.30
9.80
10.30

Percent of Consumers (%)

Total LCC in the Base Case


(weighted by market distribution)

Weighted Total LCC ($)

Figure 8.7 60W Regular AC/DC External Power Supplies: Base Case LCC Distribution
(Manufacturer Data)

8-27

8%
7%
6%
5%
4%
3%
2%
1%
0%

CSL 4
CSL 3

CSL 2
CSL 1

CSL 0

4.20
4.70
5.10
5.60
6.10
6.50
7.00
7.40
7.90
8.40
8.80
9.30
9.70
10.20
10.70
11.10
11.60

Percent of Consumers (%)

Total LCC in the Base Case


(weighted by market distribution)

Weighted Total LCC ($)

Figure 8.8 120W Regular AC/DC External Power Supplies: Base Case LCC Distribution
(Manufacturer Data)
8.4.2.2 2.5W Regular AC/DC External Power Supplies (Manufacturer Data)
Figure 8.9 and Figure 8.10 show the distribution of LCC savings and PBPs for each CSL.
In these figures, the ends of the box represent the 25th and 75th percentiles in the distribution,
while the whiskers represent the 5th and 95th percentiles. A percentile represents the portion of
sampled consumers that fall below a certain threshold. For example, an LCC savings value of
$1.00 at the 25th percentile means that 25-percent of the sampled consumers will save $1.00.
The bottom portion of the blue 25th percentile box and the top portion of the green 75th percentile
box represent the 25th and 75th percentile values in the distribution. The figures also indicate the
median and average value for each CSL. In the case of the PBP distribution, a small change in
operating cost can occasionally result in a very large PBP. As a result, the upper whisker may
become disproportionately long, skewing the average PBP value. For this reason, the median
PBP is a better estimate of the actual PBP.

8-28

Life-Cycle Cost Savings (2009$)

LCC Savings Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

1.20
1.00
0.80
0.60
0.40
0.20
0.00
-0.20
-0.40
-0.60
-0.80

Median

Average

75th Percentile
25th Percentile

CSL 1

CSL 2

CSL 3

CSL 4

Figure 8.9 2.5W Regular AC/DC External Power Supplies: Distribution of Life-Cycle Cost

Impacts (Manufacturer Data)

PBP Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

Payback Period (years)

30.0
25.0

20.0
15.0

Median

Average

10.0

75th Percentile
25th Percentile

5.0
0.0

CSL 1

CSL 2

CSL 3

CSL 4

Figure 8.10 2.5W Regular AC/DC External Power Supplies: Distribution of PBPs
(Manufacturer Data)

8-29

Table 8.4.1 summarizes the LCC and PBP results for the 2.5W EPS. As mentioned
earlier, for some consumers, DOE assigned products in the base case that are more energy
efficient than some of the energy levels under consideration. For that reason, the average LCC
impacts are not equal to the difference between the LCC of a specific energy efficiency level and
the LCC of the baseline product. Similarly with regard to PBPs shown below, DOE determined
the median and average values by excluding the percentage of consumers not impacted by a
standard at a given efficiency level. The values for average operating cost in the table are
discounted sums of the annual operating costs over the product lifetime.
For 2.5W EPSs, CSL 1 has the highest average LCC savings and a median PBP of 0.4
years. The majority of consumers (96.9-percent) are either not impacted by standard set at this
level or benefit from this standard level, while only 3.1-percent of consumers experience a net
cost. Each CSL has positive LCC savings, including CSL 4. Consumers are said to have no
impact if the base case forecast product assigned to them has a greater efficiency than the level
indicated by a standard.
Table 8.4.1 LCC and PBP Results for 2.5W Regular AC/DC External Power Supplies
(Manufacturer Data)
Life-Cycle Cost
No
Avg.
Avg.
Load
Installed Operating
Avg.
Eff.* Power
Price
Cost LCC
CSL
%
W
2009$
2009$ 2009$
0
58.3
0.500
0.00
1.73
1.73
1
67.9
0.300
0.06
1.12
1.19
2
71.0
0.130
0.35
0.90
1.24
3
73.5
0.100
0.47
0.78
1.25
4
74.0
0.053
0.63
0.74
1.36
* Eff. stands for efficiency level.
Based on an incremental MSP over the baseline.

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings Cost Impact Benefit
2009$
%
%
%
0.19
3.1
66.0
30.9
0.14 40.5
15.3
44.2
0.13 44.2
3.2
52.6
0.02 64.3
0.0
35.7

Payback Period

Median
yrs.
0.4
3.6
3.7
4.8

Mean
yrs.
1.3
12.7
12.7
17.2

8.4.2.3 18W Regular AC/DC External Power Supplies (Manufacturer Data)


Figure 8.11 and Figure 8.12 show the distribution of LCC savings and PBPs for each
CSL. In these figures, the ends of the box represent the 25th and 75th percentiles in the
distribution, while the whiskers represent the 5th and 95th percentiles. A percentile represents the
portion of sampled consumers that fall below a certain threshold. For example, an LCC savings
value of $1.00 at the 25th percentile means that 25-percent of the sampled consumers will save
$1.00. The bottom portion of the blue 25th percentile box and the top portion of the green 75th
percentile box represent the 25th and 75th percentile values in the distribution. The figures also
indicate the median and average value for each CSL. In the case of the PBP distribution, a small
change in operating cost can occasionally result in a very large PBP. As a result, the upper
whisker may become disproportionately long, skewing the average PBP value. For this reason,
the median PBP is a better estimate of the actual PBP.

8-30

Life-Cycle Cost Savings (2009$)

LCC Savings Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

5.00
4.00

3.00
2.00
Median

1.00

Average

0.00

75th

Percentile

25th Percentile

-1.00
-2.00

CSL 1

CSL 2

CSL 3

Figure 8.11 18W Regular AC/DC External Power Supplies: Distribution of Life-Cycle

Cost Impacts (Manufacturer Data)

PBP Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

45.0
Payback Period (years)

40.0
35.0

30.0
25.0
Median

20.0

Average

15.0

75th Percentile

10.0

25th Percentile

5.0
0.0

CSL 1

CSL 2

CSL 3

Figure 8.12 18W Regular AC/DC External Power Supplies: Distribution of PBPs
(Manufacturer Data)

8-31

Table 8.4.2 summarizes the LCC and PBP results for the 18W EPS. As mentioned
earlier, for some consumers, DOE assigned products in the base case that are more energy
efficient than some of the energy levels under consideration. For that reason, the average LCC
impacts are not equal to the difference between the LCC of a specific energy efficiency level and
the LCC of the baseline product. Similarly with regard to PBPs shown below, DOE determined
the median and average values by excluding the percentage of consumers not impacted by a
standard at a given efficiency level. The values for average operating cost in the table are
discounted sums of the annual operating costs over the product lifetime.
For 18W EPSs, CSL 3 has the highest average LCC savings and a median PBP of 1.5
years. The majority of consumers (81.8-percent) are either not impacted by a standard set at this
level (3.3-percent) or experience a net benefit at this level (78.5-percent). The maximum
efficiency level considered, CSL 4, also has positive LCC savings and a median PBP of 2.1
years. Consumers are said to have no impact if the base case forecast product assigned to them
has a greater efficiency than the level indicated by a standard.
Table 8.4.2 LCC and PBP Results for 18W Regular AC/DC External Power Supplies
(Manufacturer Data)
Life-Cycle Cost
No
Avg.
Avg.
Load
Installed Operating
Avg.
Eff.* Power
Price
Cost LCC
CSL
%
W
2009$
2009$ 2009$
0
76.0
0.500
0.00
6.54
6.54
1
80.3
0.300
0.44
4.99
5.43
2
83.0
0.200
0.59
4.11
4.70
3
85.4
0.100
1.11
3.35
4.45
4
87.5
0.062
1.73
2.79
4.52
* Eff. stands for efficiency level.
Based on an incremental MSP over the baseline.

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings Cost Impact Benefit
2009$
%
%
%
0.33
6.2
65.9
27.9
0.96 11.5
15.3
73.2
1.20 18.2
3.3
78.5
1.13 33.5
0.0
66.5

Payback Period

Median
yrs.
1.1
0.8
1.5
2.1

Mean
yrs.
4.9
2.8
5.5
8.0

8.4.2.4 60W Regular AC/DC External Power Supplies (Manufacturer Data)


Figure 8.13 and Figure 8.14 show the distribution of LCC savings and PBPs for each
CSL. In these figures, the ends of the box represent the 25th and 75th percentiles in the
distribution, while the whiskers represent the 5th and 95th percentiles. A percentile represents the
portion of sampled consumers that fall below a certain threshold. For example, an LCC savings
value of $1.00 at the 25th percentile means that 25-percent of the sampled consumers will save
$1.00. The bottom portion of the blue 25th percentile box and the top portion of the green 75th
percentile box represent the 25th and 75th percentile values in the distribution. The figures also
indicate the median and average value for each CSL. In the case of the PBP distribution, a small
change in operating cost can occasionally result in a very large PBP. As a result, the upper
whisker may become disproportionately long, skewing the average PBP value. For this reason,
the median PBP is a better estimate of the actual PBP.

8-32

Life-Cycle Cost Savings (2009$)

LCC Savings Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

1.00
0.50
0.00
-0.50
-1.00

Median

-1.50

Average

75th Percentile

-2.00

25th Percentile

-2.50
-3.00

CSL 1

CSL 2

CSL 3

Figure 8.13 60W Regular AC/DC External Power Supplies: Distribution of Life-Cycle

Cost Impacts (Manufacturer Data)

PBP Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

Payback Period (years)

12.0
10.0

8.0
6.0

Median

Average

4.0

75th Percentile

2.0

25th Percentile

0.0

CSL 1

CSL 2

CSL 3

Figure 8.14 60W Regular AC/DC External Power Supplies: Distribution of PBPs
(Manufacturer Data)

8-33

Table 8.4.3 summarizes the LCC and PBP results for the 60W EPS. As mentioned
earlier, for some consumers, DOE assigned products in the base case that are more energy
efficient than some of the energy levels under consideration. For that reason, the average LCC
impacts are not equal to the difference between the LCC of a specific energy efficiency level and
the LCC of the baseline product. Similarly with regard to PBPs shown below, DOE determined
the median and average values by excluding the percentage of consumers not impacted by a
standard at a given efficiency level. The values for average operating cost in the table are
discounted sums of the annual operating costs over the product lifetime.
For 60W EPSs, CSL 2 has the highest average LCC savings and a median PBP of 3.6
years. At this CSL, approximately 50.0-percent of consumer experience a net cost, but the
remaining 50.0-percent either experience no impact (15.3-percent) or a net benefit (34.7-percent)
by a standard set at this level. Consumers are said to have no impact if the base case forecast
product assigned to them has a greater efficiency than the level indicated by a standard.
Table 8.4.3 LCC and PBP Results for 60W Regular AC/DC External Power Supplies
(Manufacturer Data)
Life-Cycle Cost
No
Avg.
Avg.
Load
Installed Operating
Avg.
Eff.* Power
Price
Cost LCC
CSL
%
W
2009$
2009$ 2009$
0
85.0
0.500
0.00
4.63
4.63
1
87.0
0.500
0.34
4.08
4.43
2
87.0
0.200
1.02
3.43
4.46
3
88.0
0.073
2.28
2.90
5.18
4
91.0
0.073
4.18
2.13
6.31
* Eff. stands for efficiency level.
Based on an incremental MSP over the baseline.

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings
Cost
Impact Benefit
2009$
%
%
%
0.07
4.8
66.0
29.2
0.05
50.0
15.3
34.7
-0.65
88.5
3.3
8.2
-1.78
96.7
0.0
3.3

Payback Period

Median
yrs.
2.4
3.6
5.8
7.2

8.4.2.5 120W Regular AC/DC External Power Supplies (Manufacturer Data)


Figure 8.15 and Figure 8.16 show the distribution of LCC savings and PBPs for each
CSL. In these figures, the ends of the box represent the 25th and 75th percentiles in the
distribution, while the whiskers represent the 5th and 95th percentiles. A percentile represents the
portion of sampled consumers that fall below a certain threshold. For example, an LCC savings
value of $1.00 at the 25th percentile means that 25-percent of the sampled consumers will save
$1.00. The bottom portion of the blue 25th percentile box and the top portion of the green 75th
percentile box represent the 25th and 75th percentile values in the distribution. The figures also
indicate the median and average value for each CSL. In the case of the PBP distribution, a small
change in operating cost can occasionally result in a very large PBP. As a result, the upper
whisker may become disproportionately long, skewing the average PBP value. For this reason,
the median PBP is a better estimate of the actual PBP.

8-34

Mean
yrs.
2.4
3.6
6.0
7.2

Life-Cycle Cost Savings (2009$)

LCC Savings Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

2.00
1.00

0.00
-1.00
Median

-2.00

Average

-3.00

75th Percentile

-4.00

25th Percentile

-5.00

CSL 1

CSL 2

CSL 3

Figure 8.15 120W Regular AC/DC External Power Supplies: Distribution of Life-Cycle

Cost Impacts (Manufacturer Data)

PBP Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

Payback Period (years)

14.0
12.0

10.0
8.0
Median

6.0

Average

4.0

75th Percentile

2.0

25th Percentile

0.0

CSL 1

CSL 2

CSL 3

Figure 8.16 120W Regular AC/DC External Power Supplies: Distribution of PBPs
(Manufacturer Data)

8-35

Table 8.4.4 summarizes the LCC and PBP results for the 120W EPS. As mentioned
earlier, for some consumers, DOE assigned products in the base case that are more energy
efficient than some of the energy levels under consideration. For that reason, the average LCC
impacts are not equal to the difference between the LCC of a specific energy efficiency level and
the LCC of the baseline product. Similarly with regard to PBPs shown below, DOE determined
the median and average values by excluding the percentage of consumers not impacted by a
standard at a given efficiency level. The values for average operating cost in the table are
discounted sums of the annual operating costs over the product lifetime.
For 120W EPSs, CSL 2 has the highest average LCC savings and a median PBP of 3.0
years. The majority of consumers (55.3-percent) either experience no impact (15.2-percent) or a
net benefit (40.1-percent) from a standard set at this level. CSL 3 also has positive LCC savings,
and a median PBP of 3.3 years. Consumers are said to have no impact if the base case forecast
product assigned to them has a greater efficiency than the level indicated by a standard.
Table 8.4.4 LCC and PBP Results for 120W Regular AC/DC External Power Supplies
(Manufacturer Data)
Life-Cycle Cost
No
Avg.
Avg.
Load
Installed Operating
Avg.
Eff.* Power
Price
Cost LCC
CSL
%
W
2009$
2009$ 2009$
0
85.0
0.500
0.00
6.48
6.48
1
87.0
0.500
0.92
5.62
6.54
2
88.0
0.230
1.72
4.73
6.45
3
88.4
0.210
1.97
4.53
6.50
4
91.7
0.165
7.05
3.15
10.20
* Eff. stands for efficiency level.
Based on an incremental MSP over the baseline.

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings
Cost
Impact Benefit
2009$
%
%
%
-0.02
24.7
66.1
9.2
0.05
44.7
15.2
40.1
0.01
63.1
3.2
33.7
-3.69
99.8
0.0
0.2

Payback Period

Median
yrs.
3.6
3.0
3.3
7.6

8.4.3 Battery Charger Results (Representative Units)


DOE conducted a life-cycle cost analysis on six representative units for BCs at different
combinations of battery energy and battery voltage, one for each of the six representative product
classes: low energy, inductive charge; low energy, low voltage; low energy, high voltage;
medium energy, low voltage; high energy; and low energy, AC out. As discussed in chapter 5,
when possible, DOE created two sets of cost-efficiency curves for each of the six representative
units using data from manufacturer interviews and data from DOEs own tests/teardowns. For
some representative units, however, insufficient data was available to create a cost-efficiency
curve from tests/teardowns. In these cases, DOE only presents a single cost-efficiency curve
from its appropriate source.
Unlike the EPSs discussed previously, the MSPs for each efficiency level considered in
the BC analysis represent the total MSP attributable to a product, not the incremental MSP over a
baseline level. As such, the average installed prices can be interpreted as the total installed cost
at each CSL, and the total life-cycle cost is representative of the actual cost associated with the
product. The LCC savings, PBP calculations, and percentage of consumers with a net cost or net
benefit are unaffected by the type of MSP used, and thus remain accurate.
8-36

Mean
yrs.
3.4
2.9
3.2
7.8

8.4.3.1 Base Case LCC Distributions (Manufacturer Data)


Figure 8.17 through Figure 8.22 show the frequency charts for the base case LCC for the
six BC representative units using manufacturer data. Since the base case considers the different
efficiency levels at which consumers currently purchase products, the base case LCC distribution
is composed of several standards-levels. If all consumers purchased products at CSL 0 in the
base case, then the LCC distribution would be composed entirely of the LCC at CSL 0.
However, the LCC distribution in the base case shows the LCC results in proportion to how
many users currently purchase products at each standards-level (CSLs 0 through 3).
To find the appropriate distribution of LCCs in the base case, the total LCC was
calculated for each standards-level considered (CSLs 0 through 3). Then these distributions were
weighted by the market shares for each standards-level in the base case, as discussed in section
8.2.10. The combined chart shows the aggregated LCC distribution across all CSLs in the base
case.

7%
6%
5%
4%
CSL 3

3%

CSL 2

2%

CSL 1

1%

CSL 0

0%
5.60
6.00
6.40
6.80
7.20
7.70
8.10
8.50
8.90
9.30
9.80
10.20
10.60
11.00
11.50
11.90
12.30

Percent of Consumers (%)

Total LCC in the Base Case


(weighted by market distribution)

Weighted Total LCC ($)

Figure 8.17 Low Energy, Inductive Battery Charger: Base Case LCC Distribution
(Manufacturer Data)

8-37

35%
30%
25%
20%
CSL 3

15%

CSL 2

10%

CSL 1

5%

CSL 0

0%
0.00
1.20
2.40
3.50
4.70
5.90
7.10
8.30
9.50
10.60
11.80
13.00
14.20
15.40
16.60
17.70
18.90

Percent of Consumers (%)

Total LCC in the Base Case


(weighted by market distribution)

Weighted Total LCC ($)

Figure 8.18 Low Energy, Low Voltage Battery Charger: Base Case LCC Distribution
(Manufacturer Data)

Percent of Consumers (%)

Total LCC in the Base Case


(weighted by market distribution)

25%
20%
15%
CSL 3

10%

CSL 2

5%

CSL 1
CSL 0

0%

Weighted Total LCC ($)

Figure 8.19 Low Energy, High Voltage Battery Charger: Base Case LCC Distribution
(Manufacturer Data)

8-38

Percent of Consumers (%)

Total LCC in the Base Case


(weighted by market distribution)

9%
8%
7%
6%
5%
4%
3%
2%
1%
0%

CSL 3
CSL 2

CSL 1
CSL 0

Weighted Total LCC ($)

Figure 8.20 Medium Energy, Low Voltage Battery Charger: Base Case LCC Distribution
(Manufacturer Data)

7%
6%
5%
4%
3%
2%
1%
0%

CSL 3
CSL 2

CSL 1
CSL 0

299.80
312.90
326.10
339.20
352.30
365.40
378.60
391.70
404.80
417.90
431.00
444.20
457.30
470.40
483.50
496.70
509.80

Percent of Consumers (%)

Total LCC in the Base Case


(weighted by market distribution)

Weighted Total LCC ($)

Figure 8.21 High Energy Battery Charger: Base Case LCC Distribution (Manufacturer

Data)

8-39

7%
6%
5%
4%
CSL 3

3%

CSL 2

2%

CSL 1

1%

CSL 0

0%
9.20
10.10
11.00
12.00
12.90
13.90
14.80
15.80
16.70
17.70
18.60
19.60
20.50
21.40
22.40
23.30
24.30

Percent of Consumers (%)

Total LCC in the Base Case


(weighted by market distribution)

Weighted Total LCC ($)

Figure 8.22 Low Energy, AC Out Battery Charger: Base Case LCC Distribution
(Manufacturer Data)
8.4.3.2 Base Case LCC Distributions (Test/Teardown Data)
Figure 8.23 through Figure 8.26 show the frequency charts for the base case LCC for four
of the BC representative units using test/teardown data. DOE only had sufficient test/teardown
data for four of the six BC representative units. Since the base case considers the different
efficiency levels at which consumers currently purchase products, the base case LCC distribution
is composed of several standards-levels. If all consumers purchased products at CSL 0 in the
base case, then the LCC distribution would be composed entirely of the LCC at CSL 0.
However, the LCC distribution in the base case shows the LCC results in proportion to how
many users currently purchase products at each standards-level (CSLs 0 through 3).
To find the appropriate distribution of LCCs in the base case, the total LCC was
calculated for each standards-level considered (CSLs 0 through 3). Then these distributions were
weighted by the market shares for each standards-level in the base case, as discussed in section
8.2.10. The combined chart shows the aggregated LCC distribution across all CSLs in the base
case.

8-40

18%
16%
14%
12%
10%
8%
6%
4%
2%
0%

CSL 3
CSL 2

CSL 1
CSL 0

1.60
2.00
2.40
2.80
3.20
3.60
4.00
4.40
4.80
5.20
5.60
6.00
6.40
6.80
7.20
7.60
8.00
8.40
8.80
9.20
9.60
10.00

Percent of Consumers (%)

Total LCC in the Base Case


(weighted by market distribution)

Weighted Total LCC ($)

Figure 8.23 Low Energy, Low Voltage Battery Charger: Base Case LCC Distribution
(Test/Teardown Data)

30%
25%
20%
15%

CSL 3

10%

CSL 2

5%

CSL 1
CSL 0

0%
8.60
10.30
12.00
13.80
15.50
17.20
18.90
20.70
22.40
24.10
25.80
27.50
29.30
31.00
32.70
34.40
36.10
37.90
39.60
41.30
43.00

Percent of Consumers (%)

Total LCC in the Base Case


(weighted by market distribution)

Weighted Total LCC ($)

Figure 8.24 Low Energy, High Voltage Battery Charger: Base Case LCC Distribution
(Test/Teardown Data)

8-41

14%
12%
10%
8%
6%
4%
2%
0%

CSL 3
CSL 2

CSL 1
CSL 0

70.20
81.90
93.60
105.30
117.00
128.70
140.40
152.10
163.80
175.50
187.30
199.00
210.70
222.40
234.10
245.80
257.50
269.20
280.90
292.60

Percent of Consumers (%)

Total LCC in the Base Case


(weighted by market distribution)

Weighted Total LCC ($)

Figure 8.25 Medium Energy, Low Voltage Battery Charger: Base Case LCC Distribution
(Test/Teardown Data)

10%
9%
8%
7%
6%
5%
4%
3%
2%
1%
0%

CSL 3
CSL 2

CSL 1
CSL 0

5.50
6.50
7.50
8.50
9.50
10.50
11.50
12.50
13.50
14.40
15.40
16.40
17.40
18.40
19.40
20.40
21.40
22.40
23.40
24.40

Percent of Consumers (%)

Total LCC in the Base Case


(weighted by market distribution)

Weighted Total LCC ($)

Figure 8.26 Low Energy, AC Out Battery Charger: Base Case LCC Distribution
(Test/Teardown Data)

8-42

8.4.3.3 Low Energy, Inductive Battery Chargers (Manufacturer Data)


Figure 8.27 and Figure 8.28 show the distribution of LCC savings and PBPs for each
CSL. In these figures, the ends of the box represent the 25th and 75th percentiles in the
distribution, while the whiskers represent the 5th and 95th percentiles. A percentile represents the
portion of sampled consumers that fall below a certain threshold. For example, an LCC savings
value of $1.00 at the 25th percentile means that 25-percent of the sampled consumers will save
$1.00. The bottom portion of the blue 25th percentile box and the top portion of the green 75th
percentile box represent the 25th and 75th percentile values in the distribution. The figures also
indicate the median and average value for each CSL. In the case of the PBP distribution, a small
change in operating cost can occasionally result in a very large PBP. As a result, the upper
whisker may become disproportionately long, skewing the average PBP value. For this reason,
the median PBP is a better estimate of the actual PBP.

Life-Cycle Cost Savings (2009$)

LCC Savings Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

6.00
5.00

4.00
3.00

Median

Average

2.00

1.00

75th

Percentile

25th

Percentile

0.00

CSL 1

CSL 2

CSL 3

Figure 8.27 Low Energy, Inductive Battery Chargers: Distribution of Life-Cycle Cost

Impacts (Manufacturer Data)

8-43

PBP Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

Payback Period (years)

1.2
1.0

0.8
0.6

Median

Average

0.4

75th

0.2

Percentile

25th Percentile

0.0

CSL 1

CSL 2

CSL 3

Figure 8.28 Low Energy, Inductive Battery Chargers: Distribution of PBPs (Manufacturer
Data)
Table 8.4.5 summarizes the LCC and PBP results for the low energy, inductive BC. As
mentioned earlier, for some consumers, DOE assigned products in the base case that are more
energy efficient than some of the energy levels under consideration. For that reason, the average
LCC impacts are not equal to the difference between the LCC of a specific energy efficiency
level and the LCC of the baseline product. Similarly with regard to PBPs shown below, DOE
determined the median and average values by excluding the percentage of consumers not
impacted by a standard at a given efficiency level. The values for average operating cost in the
table are discounted sums of the annual operating costs over the product lifetime.
For low energy, inductive BCs, CSL 3 has the highest average LCC savings and a median
PBP of 0.9 years. Every consumer (100.0-percent) experiences a net benefit at this level.
Consumers are said to have no impact if the base case forecast product assigned to them has a
greater efficiency than the level indicated by a standard.

8-44

Table 8.4.5 LCC and PBP Results for Low Energy, Inductive Battery Chargers
(Manufacturer Data)

CSL
0
1
2
3

UEC
kWh
10.0
7.0
3.5
1.5

Avg.
Installed
Price
2009$
4.39
4.61
4.91
5.12

Life-Cycle Cost
Avg.
Operating
Cost
2009$
4.81
3.37
1.68
0.72

Avg.
LCC
2009$
9.20
7.97
6.59
5.84

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings
Cost
Impact
Benefit
2009$
%
%
%
0.93
0.0
24.9
75.1
2.14
0.0
12.4
87.6
2.90
0.0
0.0
100.0

Payback Period

Median
yrs.
0.7
0.8
0.9

Mean
yrs.
0.7
0.8
0.8

8.4.3.4 Low Energy, Inductive Battery Chargers (Test/Teardown Data)


DOE did not have sufficient test/teardown data to perform an LCC analysis of the low
energy, inductive representative unit. Section 8.4.3.3 contains LCC and PBP results for this
representative unit using manufacturer data.
8.4.3.5 Low Energy, Low Voltage Battery Chargers (Manufacturer Data)
Figure 8.29 and Figure 8.30 show the distribution of LCC savings and PBPs for each
CSL. In these figures, the ends of the box represent the 25th and 75th percentiles in the
distribution, while the whiskers represent the 5th and 95th percentiles. A percentile represents the
portion of sampled consumers that fall below a certain threshold. For example, an LCC savings
value of $1.00 at the 25th percentile means that 25-percent of the sampled consumers will save
$1.00. The bottom portion of the blue 25th percentile box and the top portion of the green 75th
percentile box represent the 25th and 75th percentile values in the distribution. The figures also
indicate the median and average value for each CSL. In the case of the PBP distribution, a small
change in operating cost can occasionally result in a very large PBP. As a result, the upper
whisker may become disproportionately long, skewing the average PBP value. For this reason,
the median PBP is a better estimate of the actual PBP.

8-45

Life-Cycle Cost Savings (2009$)

LCC Savings Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

4.00
2.00
0.00
-2.00
-4.00
-6.00
-8.00
-10.00
-12.00
-14.00
-16.00
-18.00

Median

Average

75th

Percentile

25th Percentile

CSL 1

CSL 2

Figure 8.29 Low Energy, Low Voltage Battery Chargers: Distribution of Life-Cycle Cost

Impacts (Manufacturer Data)

PBP Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

180.0
Payback Period (years)

160.0
140.0

120.0
100.0
Median

80.0

Average

60.0

75th

40.0

Percentile

25th Percentile

20.0
0.0

CSL 1

CSL 2

Figure 8.30 Low Energy, Low Voltage Battery Chargers: Distribution of PBPs
(Manufacturer Data)
Table 8.4.6 summarizes the LCC and PBP results for the low energy, low voltage BC.
As mentioned earlier, for some consumers, DOE assigned products in the base case that are more
8-46

energy efficient than some of the energy levels under consideration. For that reason, the average
LCC impacts are not equal to the difference between the LCC of a specific energy efficiency
level and the LCC of the baseline product. Similarly with regard to PBPs shown below, DOE
determined the median and average values by excluding the percentage of consumers not
impacted by a standard at a given efficiency level. The values for average operating cost in the
table are discounted sums of the annual operating costs over the product lifetime.
For low energy, low voltage BCs, CSL 1 has the highest average LCC savings ($0.20)
and a median PBP of 1.5 years. The majority of consumers (60.8-percent) are not impacted by a
standard set at this level, but 32.6-percent of consumers experience a net benefit. Consumers are
said to have no impact if the base case forecast product assigned to them has a greater
efficiency than the level indicated by a standard.
Table 8.4.6 LCC and PBP Results for Low Energy, Low Voltage Battery Chargers
(Manufacturer Data)

CSL
0
1
2

UEC
kWh
10.0
6.0
1.2

Avg.
Installed
Price
2009$
1.42
2.13
18.09

Life-Cycle Cost
Avg.
Operating
Cost
2009$
3.30
2.08
0.46

Avg.
LCC
2009$
4.72
4.20
18.54

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings
Cost
Impact
Benefit
2009$
%
%
%
0.20
6.6
60.8
32.6
-13.74
97.2
2.8
0.0

Payback Period

Median
yrs.
1.5
25.4

Mean
yrs.
2.2
38.9

8.4.3.6 Low Energy, Low Voltage Battery Chargers (Test/Teardown Data)


Figure 8.31 through Figure 8.32 show the distribution of LCC savings and PBPs for each
CSL. In these figures, the ends of the box represent the 25th and 75th percentiles in the
distribution, while the whiskers represent the 5th and 95th percentiles. A percentile represents the
portion of sampled consumers that fall below a certain threshold. For example, an LCC savings
value of $1.00 at the 25th percentile means that 25-percent of the sampled consumers will save
$1.00. The bottom portion of the blue 25th percentile box and the top portion of the green 75th
percentile box represent the 25th and 75th percentile values in the distribution. The figures also
indicate the median and average value for each CSL. In the case of the PBP distribution, a small
change in operating cost can occasionally result in a very large PBP. As a result, the upper
whisker may become disproportionately long, skewing the average PBP value. For this reason,
the median PBP is a better estimate of the actual PBP.
For the low energy, low voltage representative unit, DOE did not have sufficient
test/teardown data to accurately evaluate a baseline unit (CSL 0). For this reason, the figures
show a mean and median value of $0 and 0 years for the LCC savings and PBP, respectively.
For additional analyses of the baseline CSL for this representative unit, please see section
8.4.3.5.

8-47

Life-Cycle Cost Savings (2009$)

LCC Savings Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

3.00
2.00

1.00
0.00

Median

Average

-1.00

75th

-2.00

Percentile

25th Percentile

-3.00

CSL 1

CSL 2

Figure 8.31 Low Energy, Low Voltage Battery Chargers: Distribution of Life-Cycle Cost

Impacts (Test/Teardown Data)

PBP Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

Payback Period (years)

30.0
25.0

20.0
15.0

Median

Average

10.0

75th

Percentile

25th Percentile

5.0
0.0

CSL 1

CSL 2

Figure 8.32 Low Energy, Low Voltage Battery Chargers: Distribution of PBPs
(Test/Teardown Data)
Table 8.4.7 summarizes the LCC and PBP results for the low energy, low voltage BC
using test/teardown data. As mentioned earlier, for some consumers, DOE assigned products in
8-48

the base case that are more energy efficient than some of the energy levels under consideration.
For that reason, the average LCC impacts are not equal to the difference between the LCC of a
specific energy efficiency level and the LCC of the baseline product. Similarly with regard to
PBPs shown below, DOE determined the median and average values by excluding the
percentage of consumers not impacted by a standard at a given efficiency level. The values for
average operating cost in the table are discounted sums of the annual operating costs over the
product lifetime.
For low energy, low voltage BCs, CSL 1 is evaluated as the baseline unit since DOE did
not have sufficient test/teardown data to evaluate a unit at CSL 0. The only standard level that
was evaluated was CSL 2, which has negative LCC savings of -$0.90 and a median PBP of 4.7
years. Consumers are said to have no impact if the base case forecast product assigned to them
has a greater efficiency than the level indicated by a standard.
Table 8.4.7 LCC and PBP Results for Low Energy, Low Voltage Battery Chargers
(Test/Teardown Data)

CSL
0
1
2

UEC
kWh
10.0
6.0
1.2

Avg.
Installed
Price
2009$
N/A
1.42
4.43

Life-Cycle Cost
Avg.
Operating
Cost
2009$
N/A
2.48
0.38

Avg.
LCC
2009$
N/A
3.90
4.82

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings
Cost
Impact
Benefit
2009$
%
%
%
N/A
N/A
N/A
N/A
-0.90
73.7
2.8
23.5

Payback Period

Median
yrs.
N/A
4.7

Mean
yrs.
N/A
6.7

8.4.3.7 Low Energy, High Voltage Battery Chargers (Manufacturer Data)


Figure 8.33 and Figure 8.34 show the distribution of LCC savings and PBPs for each
CSL. In these figures, the ends of the box represent the 25th and 75th percentiles in the
distribution, while the whiskers represent the 5th and 95th percentiles. A percentile represents the
portion of sampled consumers that fall below a certain threshold. For example, an LCC savings
value of $1.00 at the 25th percentile means that 25-percent of the sampled consumers will save
$1.00. The bottom portion of the blue 25th percentile box and the top portion of the green 75th
percentile box represent the 25th and 75th percentile values in the distribution. The figures also
indicate the median and average value for each CSL. At CSL 1, there is no box present because
both the 25th and 75th percentiles equal 0, represented by the median. In the case of the PBP
distribution, a small change in operating cost can occasionally result in a very large PBP. As a
result, the upper whisker may become disproportionately long, skewing the average PBP value.
For this reason, the median PBP is a better estimate of the actual PBP.

8-49

Life-Cycle Cost Savings (2009$)

LCC Savings Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

6.00
4.00

2.00
0.00

Median

Average

-2.00

75th

-4.00

Percentile

25th Percentile

-6.00

CSL 1

CSL 2

Figure 8.33 Low Energy, High Voltage BC: Distribution of Life-Cycle Cost Impacts
(Manufacturer Data)

Payback Period (years)

PBP Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

20.0
18.0
16.0
14.0
12.0
10.0
8.0
6.0
4.0
2.0
0.0

Median

Average

CSL 1

75th

Percentile

25th

Percentile

CSL 2

Figure 8.34 Low Energy, High Voltage BC: Distribution of PBPs (Manufacturer Data)
Table 8.4.8 summarizes the LCC and PBP results for the low energy, high voltage BC.
As mentioned earlier, for some consumers, DOE assigned products in the base case that are more
energy efficient than some of the energy levels under consideration. For that reason, the average
8-50

LCC impacts are not equal to the difference between the LCC of a specific energy efficiency
level and the LCC of the baseline product. Similarly with regard to PBPs shown below, DOE
determined the median and average values by excluding the percentage of consumers not
impacted by a standard at a given efficiency level. The values for average operating cost in the
table are discounted sums of the annual operating costs over the product lifetime.
For low energy, high voltage BCs, CSL 1 has the highest average LCC savings ($0.97)
and a median PBP of 3.4 years. The majority of consumers (81.3-percent) are not impacted by a
standard set at this level. Consumers are said to have no impact if the base case forecast
product assigned to them has a greater efficiency than the level indicated by a standard.
Table 8.4.8 LCC and PBP Results for Low Energy, High Voltage Battery Chargers
(Manufacturer Data)

CSL
0
1
2

UEC
kWh
39.4
10.5
6.1

Avg.
Installed
Price
2009$
8.04
16.13
20.60

Life-Cycle Cost
Avg.
Operating
Cost
2009$
16.65
4.31
2.48

Avg.
LCC
2009$
24.69
20.44
23.08

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings
Cost
Impact
Benefit
2009$
%
%
%
0.97
9.0
81.3
9.7
-0.65
53.8
38.3
7.9

Payback Period

Median
yrs.
3.4
9.3

Mean
yrs.
3.1
9.1

8.4.3.8 Low Energy, High Voltage Battery Chargers (Test/Teardown Data)


Figure 8.35 and Figure 8.36 show the distribution of LCC savings and PBPs for each
CSL. In these figures, the ends of the box represent the 25th and 75th percentiles in the
distribution, while the whiskers represent the 5th and 95th percentiles. A percentile represents the
portion of sampled consumers that fall below a certain threshold. For example, an LCC savings
value of $1.00 at the 25th percentile means that 25-percent of the sampled consumers will save
$1.00. The bottom portion of the blue 25th percentile box and the top portion of the green 75th
percentile box represent the 25th and 75th percentile values in the distribution. The figures also
indicate the median and average value for each CSL. At CSL 1, there is no box present because
both the 25th and 75th percentiles equal 0, represented by the median. In the case of the PBP
distribution, a small change in operating cost can occasionally result in a very large PBP. As a
result, the upper whisker may become disproportionately long, skewing the average PBP value.
For this reason, the median PBP is a better estimate of the actual PBP.

8-51

Life-Cycle Cost Savings (2009$)

LCC Savings Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

10.00
8.00

6.00
4.00

Median

Average

2.00

75th

0.00

Percentile

25th Percentile

-2.00

CSL 1

CSL 2

Figure 8.35 Low Energy, High Voltage BC: Distribution of Life-Cycle Cost Impacts
(Test/Teardown Data)

PBP Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

Payback Period (years)

6.0
5.0

4.0
3.0

Median

Average

2.0

75th

Percentile

25th Percentile

1.0
0.0

CSL 1

CSL 2

Figure 8.36 Low Energy, High Voltage BC: Distribution of PBPs (Test/Teardown Data)
Table 8.4.9 summarizes the LCC and PBP results for the low energy, high voltage BC
using the test/teardown data. As mentioned earlier, for some consumers, DOE assigned products
in the base case that are more energy efficient than some of the energy levels under
8-52

consideration. For that reason, the average LCC impacts are not equal to the difference between
the LCC of a specific energy efficiency level and the LCC of the baseline product. Similarly
with regard to PBPs shown below, DOE determined the median and average values by excluding
the percentage of consumers not impacted by a standard at a given efficiency level. The values
for average operating cost in the table are discounted sums of the annual operating costs over the
product lifetime.
For low energy, high voltage BCs, CSL 2 has the highest average LCC savings ($1.74)
and a median PBP of 1.5 years. Over half of all consumers (50.7-percent) experience a net
benefit from a standard set at this level, and an additional 38.2-percent of consumers already
purchase products at CSL 2 and would therefore experience no impact from a standard at this
level. Consumers are said to have no impact if the base case forecast product assigned to them
has a greater efficiency than the level indicated by a standard.
Table 8.4.9 LCC and PBP Results for Low Energy, High Voltage Battery Chargers
(Test/Teardown)

CSL
0
1
2

UEC
kWh
39.4
10.5
6.1

Avg.
Installed
Price
2009$
8.04
12.23
13.20

Life-Cycle Cost
Avg.
Operating
Cost
2009$
15.38
4.14
2.52

Avg.
LCC
2009$
23.42
16.38
15.72

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings
Cost
Impact
Benefit
2009$
%
%
%
1.33
0.4
81.5
18.1
1.74
11.1
38.2
50.7

Payback Period

Median
yrs.
1.3
1.5

Mean
yrs.
1.6
2.5

8.4.3.9 Medium Energy, Low Voltage Battery Chargers (Manufacturer Data)


Figure 8.37 and Figure 8.38 show the distribution of LCC savings and PBPs for each
CSL. In these figures, the ends of the box represent the 25th and 75th percentiles in the
distribution, while the whiskers represent the 5th and 95th percentiles. A percentile represents the
portion of sampled consumers that fall below a certain threshold. For example, an LCC savings
value of $1.00 at the 25th percentile means that 25-percent of the sampled consumers will save
$1.00. The bottom portion of the blue 25th percentile box and the top portion of the green 75th
percentile box represent the 25th and 75th percentile values in the distribution. The figures also
indicate the median and average value for each CSL. In the case of the PBP distribution, a small
change in operating cost can occasionally result in a very large PBP. As a result, the upper
whisker may become disproportionately long, skewing the average PBP value. For this reason,
the median PBP is a better estimate of the actual PBP.

8-53

Life-Cycle Cost Savings (2009$)

LCC Savings Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

40.00
20.00
0.00

-20.00
-40.00
Median

-60.00

Average

-80.00

75th

-100.00

Percentile

25th Percentile

-120.00
-140.00

CSL 1

CSL 2

CSL 3

Figure 8.37 Medium Energy, Low Voltage Battery Chargers: Distribution of Life-Cycle

Cost Impacts (Manufacturer Data)

PBP Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

Payback Period (years)

300.0
250.0

200.0
150.0

Median

Average

100.0

75th

Percentile

25th Percentile

50.0
0.0

CSL 1

CSL 2

CSL 3

Figure 8.38 Medium Energy, Low Voltage Battery Chargers: Distribution of PBPs
(Manufacturer Data)
Table 8.4.10 summarizes the LCC and PBP results for the medium energy, low voltage
BC. As mentioned earlier, for some consumers, DOE assigned products in the base case that are
8-54

more energy efficient than some of the energy levels under consideration. For that reason, the
average LCC impacts are not equal to the difference between the LCC of a specific energy
efficiency level and the LCC of the baseline product. Similarly with regard to PBPs shown
below, DOE determined the median and average values by excluding the percentage of
consumers not impacted by a standard at a given efficiency level. The values for average
operating cost in the table are discounted sums of the annual operating costs over the product
lifetime.
For medium energy, low voltage BCs, CSL 1 has the highest average LCC savings
($2.08) and a median PBP of 2.4 years. Over half of all consumers (56.9-percent) are not
impacted by a standard set at this level, and 33.1-percent of consumers experience a net benefit
at this level. Consumers are said to have no impact if the base case forecast product assigned
to them has a greater efficiency than the level indicated by a standard.
Table 8.4.10 LCC and PBP Results for Medium Energy, Low Voltage Battery Chargers
(Manufacturer Data)

CSL
0
1
2
3

UEC
kWh
202.7
159.6
100.0
75.0

Avg.
Installed
Price
2009$
45.91
59.09
112.93
205.20

8.4.3.10
Data)

Life-Cycle Cost
Avg.
Operating
Cost
2009$
90.58
72.76
40.69
30.43

Avg.
LCC
2009$
136.49
131.85
153.62
235.63

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings
Cost
Impact
Benefit
2009$
%
%
%
2.08
10.0
56.9
33.1
-11.07
48.4
40.5
11.1
-93.07
99.9
0.0
0.1

Payback Period

Median
yrs.
2.4
6.4
17.2

Mean
yrs.
8.1
7.7
42.2

Medium Energy, Low Voltage Battery Chargers (Test/Teardown

Figure 8.39 and Figure 8.40 show the distribution of LCC savings and PBPs for each
CSL. In these figures, the ends of the box represent the 25th and 75th percentiles in the
distribution, while the whiskers represent the 5th and 95th percentiles. A percentile represents the
portion of sampled consumers that fall below a certain threshold. For example, an LCC savings
value of $1.00 at the 25th percentile means that 25-percent of the sampled consumers will save
$1.00. The bottom portion of the blue 25th percentile box and the top portion of the green 75th
percentile box represent the 25th and 75th percentile values in the distribution. The figures also
indicate the median and average value for each CSL. In the case of the PBP distribution, a small
change in operating cost can occasionally result in a very large PBP. As a result, the upper
whisker may become disproportionately long, skewing the average PBP value. For this reason,
the median PBP is a better estimate of the actual PBP.

8-55

Life-Cycle Cost Savings (2009$)

LCC Savings Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

70.00
60.00

50.00
40.00
Median

30.00

Average

20.00

75th

10.00

25th Percentile

Percentile

0.00

CSL 1
Figure 8.39 Medium Energy, Low Voltage Battery Chargers: Distribution of Life-Cycle

Cost Impacts (Test/Teardown Data)

PBP Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

Payback Period (years)

1.4
1.2

1.0
0.8
Median

0.6

Average

0.4

75th

0.2

25th Percentile

Percentile

0.0

CSL 1
Figure 8.40 Medium Energy, Low Voltage Battery Chargers: Distribution of PBPs
(Test/Teardown Data)
Table 8.4.11 summarizes the LCC and PBP results for the medium energy, low voltage
BC using the test/teardown data. As mentioned earlier, for some consumers, DOE assigned
8-56

products in the base case that are more energy efficient than some of the energy levels under
consideration. For that reason, the average LCC impacts are not equal to the difference between
the LCC of a specific energy efficiency level and the LCC of the baseline product. Similarly
with regard to PBPs shown below, DOE determined the median and average values by excluding
the percentage of consumers not impacted by a standard at a given efficiency level. The values
for average operating cost in the table are discounted sums of the annual operating costs over the
product lifetime.
For medium energy, low voltage BCs using test/teardown data, CSL 1 is the only
standards-level above the baseline that DOE had sufficient test/teardown data to evaluate. CSL 1
has positive average LCC savings ($11.20) and a median PBP of 0.9 years. The majority of
consumers (56.2-percent) already purchase products at CSL 1, but all consumers impacted by a
standard set at CSL 1 will experience a net benefit. Consumers are said to have no impact if
the base case forecast product assigned to them has a greater efficiency than the level indicated
by a standard.
Table 8.4.11 LCC and PBP Results for Medium Energy, Low Voltage Battery Chargers
(Test/Teardown Data)

CSL
0
1

UEC
kWh
202.7
159.6

Avg.
Installed
Price
2009$
45.91
50.64

8.4.3.11

Life-Cycle Cost
Avg.
Operating
Cost
2009$
112.90
83.11

Avg.
LCC
2009$
158.81
133.75

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings
Cost
Impact
Benefit
2009$
%
%
%
11.20
0.0
56.2
43.8

Payback Period

Median
yrs.
0.9

Mean
yrs.
0.9

High Energy Battery Chargers (Manufacturer Data)

Figure 8.41 and Figure 8.42 show the distribution of LCC savings and PBPs for each
CSL. In these figures, the ends of the box represent the 25th and 75th percentiles in the
distribution, while the whiskers represent the 5th and 95th percentiles. A percentile represents the
portion of sampled consumers that fall below a certain threshold. For example, an LCC savings
value of $1.00 at the 25th percentile means that 25-percent of the sampled consumers will save
$1.00. The bottom portion of the blue 25th percentile box and the top portion of the green 75th
percentile box represent the 25th and 75th percentile values in the distribution. The figures also
indicate the median and average value for each CSL. In the case of the PBP distribution, a small
change in operating cost can occasionally result in a very large PBP. As a result, the upper
whisker may become disproportionately long, skewing the average PBP value. For this reason,
the median PBP is a better estimate of the actual PBP.

8-57

Life-Cycle Cost Savings (2009$)

LCC Savings Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

20.00
0.00

-20.00
-40.00
Median

-60.00

Average

-80.00

75th

Percentile

25th Percentile

-100.00
-120.00

CSL 1

CSL 2

CSL 3

Figure 8.41 High Energy Battery Chargers: Distribution of Life-Cycle Cost Impacts
(Manufacturer Data)

PBP Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

Payback Period (years)

25.0
20.0
15.0
Median

10.0

Average

75th
5.0

Percentile

25th Percentile

0.0

CSL 1

CSL 2

CSL 3

Figure 8.42 High Energy Battery Chargers: Distribution of PBPs (Manufacturer Data)
Table 8.4.12 summarizes the LCC and PBP results for the high energy BC. As
mentioned earlier, for some consumers, DOE assigned products in the base case that are more
energy efficient than some of the energy levels under consideration. For that reason, the average
8-58

LCC impacts are not equal to the difference between the LCC of a specific energy efficiency
level and the LCC of the baseline product. Similarly with regard to PBPs shown below, DOE
determined the median and average values by excluding the percentage of consumers not
impacted by a standard at a given efficiency level. The values for average operating cost in the
table are discounted sums of the annual operating costs over the product lifetime.
For high energy BCs, CSL 1 has the highest average LCC savings ($0.10) and a median
PBP of 5.8 years. Half of all consumers (50.0-percent) are not impacted by a standard set at this
level, and 20.3-percent of consumers experience a net benefit at this level. Consumers are said
to have no impact if the base case forecast product assigned to them has a greater efficiency
than the level indicated by a standard.
Table 8.4.12 LCC and PBP Results for High Energy Battery Chargers (Manufacturer
Data)

CSL
0
1
2
3

UEC
kWh
290.0
250.0
200.0
150.0

Avg.
Installed
Price
2009$
195.80
219.13
266.16
361.77

8.4.3.12

Life-Cycle Cost
Avg.
Operating
Cost
2009$
171.85
148.15
118.52
88.89

Avg.
LCC
2009$
367.65
367.28
384.68
450.66

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings
Cost Impact
Benefit
2009$
%
%
%
0.10
29.7
50.0
20.3
-12.07
64.2
30.1
5.7
-78.04 100.0
0.0
0.0

Payback Period

Median
yrs.
5.8
7.8
12.6

Mean
yrs.
5.4
7.7
13.6

High Energy Battery Chargers (Test/Teardown Data)

DOE did not have sufficient test/teardown data to perform an LCC analysis of the low
energy, inductive representative unit. Section 8.4.3.11 contains LCC and PBP results for this
representative unit using manufacturer data.
8.4.3.13

Low Energy, AC Out Battery Chargers (Manufacturer Data)

Figure 8.43 and Figure 8.44 show the distribution of LCC savings and PBPs for each
CSL. In these figures, the ends of the box represent the 25th and 75th percentiles in the
distribution, while the whiskers represent the 5th and 95th percentiles. A percentile represents the
portion of sampled consumers that fall below a certain threshold. For example, an LCC savings
value of $1.00 at the 25th percentile means that 25-percent of the sampled consumers will save
$1.00. The bottom portion of the blue 25th percentile box and the top portion of the green 75th
percentile box represent the 25th and 75th percentile values in the distribution. The figures also
indicate the median and average value for each CSL. In the case of the PBP distribution, a small
change in operating cost can occasionally result in a very large PBP. As a result, the upper
whisker may become disproportionately long, skewing the average PBP value. For this reason,
the median PBP is a better estimate of the actual PBP.

8-59

Life-Cycle Cost Savings (2009$)

LCC Savings Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

14.00
12.00

10.00
8.00
Median

6.00

Average

4.00

75th

2.00

25th Percentile

Percentile

0.00

CSL 1

CSL 2

CSL 3

Figure 8.43 Low Energy, AC Out Battery Chargers: Distribution of Life-Cycle Cost

Impacts (Manufacturer Data)

PBP Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

Payback Period (years)

2.5
2.0
1.5
Median

1.0

Average

75th
0.5

Percentile

25th Percentile

0.0

CSL 1

CSL 2

CSL 3

Figure 8.44 Low Energy, AC Out Battery Chargers: Distribution of PBPs (Manufacturer
Data)
Table 8.4.13 summarizes the LCC and PBP results for the low energy, AC out BC. As
mentioned earlier, for some consumers, DOE assigned products in the base case that are more
8-60

energy efficient than some of the energy levels under consideration. For that reason, the average
LCC impacts are not equal to the difference between the LCC of a specific energy efficiency
level and the LCC of the baseline product. Similarly with regard to PBPs shown below, DOE
determined the median and average values by excluding the percentage of consumers not
impacted by a standard at a given efficiency level. The values for average operating cost in the
table are discounted sums of the annual operating costs over the product lifetime.
For low energy, AC out BCs, CSL 3 has the highest average LCC savings ($6.88) and a
median PBP of 1.5 years. All consumers (100.0-percent) experience a net benefit from a
standard set at this level. Consumers are said to have no impact if the base case forecast
product assigned to them has a greater efficiency than the level indicated by a standard.
Table 8.4.13 LCC and PBP Results for Low Energy, AC Out Battery Chargers
(Manufacturer Data)

CSL
0
1
2
3

UEC
kWh
19.6
6.4
4.0
1.5

Avg.
Installed
Price
2009$
5.93
7.62
8.08
8.65

8.4.3.14

Life-Cycle Cost
Avg.
Operating
Cost
2009$
11.07
3.60
2.26
0.85

Avg.
LCC
2009$
17.00
11.22
10.34
9.50

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings
Cost
Impact
Benefit
2009$
%
%
%
5.16
0.0
11.0
89.0
6.04
0.0
0.0
100.0
6.88
0.0
0.0
100.0

Payback Period

Median
yrs.
1.3
1.4
1.5

Mean
yrs.
1.2
1.3
1.5

Low Energy, AC Out Battery Chargers (Test/Teardown Data)

Figure 8.45 and Figure 8.46 show the distribution of LCC savings and PBPs for each
CSL. In these figures, the ends of the box represent the 25th and 75th percentiles in the
distribution, while the whiskers represent the 5th and 95th percentiles. A percentile represents the
portion of sampled consumers that fall below a certain threshold. For example, an LCC savings
value of $1.00 at the 25th percentile means that 25-percent of the sampled consumers will save
$1.00. The bottom portion of the blue 25th percentile box and the top portion of the green 75th
percentile box represent the 25th and 75th percentile values in the distribution. The figures also
indicate the median and average value for each CSL. In the case of the PBP distribution, a small
change in operating cost can occasionally result in a very large PBP. As a result, the upper
whisker may become disproportionately long, skewing the average PBP value. For this reason,
the median PBP is a better estimate of the actual PBP.
The low energy, AC out BC had test/teardown data that indicated a negative-sloping costefficiency curve. As a result, all consumers benefit from standards. This is seen in Figure 8.45,
where the entire distribution of LCC savings is positive. Similarly, all consumers experience an
immediate payback, so Figure 8.46 shows the median and average payback period as 0 years
with no distribution of values around the average.

8-61

Life-Cycle Cost Savings (2009$)

LCC Savings Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

14.00
12.00

10.00
8.00
Median

6.00

Average

4.00

75th

2.00

25th Percentile

Percentile

0.00

CSL 1
Figure 8.45 Low Energy, AC Out Battery Chargers: Distribution of Life-Cycle Cost

Impacts (Test/Teardown Data)

Payback Period (years)

PBP Ranges by Efficiency Level


Box 25% - 75%; Whisker: 5% - 95%

1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0

Median

Average

75th

Percentile

25th Percentile

CSL 1
Figure 8.46 Low Energy, AC Out Battery Chargers: Distribution of PBPs (Test/Teardown
Data)
Table 8.4.14 summarizes the LCC and PBP results for the low energy, AC out BC using
test/teardown data. As mentioned earlier, for some consumers, DOE assigned products in the
8-62

base case that are more energy efficient than some of the energy levels under consideration. For
that reason, the average LCC impacts are not equal to the difference between the LCC of a
specific energy efficiency level and the LCC of the baseline product. Similarly with regard to
PBPs shown below, DOE determined the median and average values by excluding the
percentage of consumers not impacted by a standard at a given efficiency level. The values for
average operating cost in the table are discounted sums of the annual operating costs over the
product lifetime.
As mentioned previously, DOEs cost estimates for low energy, AC out BCs decrease
from the baseline unit, producing a negative-sloping cost-efficiency curve. As a result, LCC
savings occur automatically for any user switching from a baseline product. Similarly, all of
these consumers will experience an instantaneous payback since the purchase price of a product
at CSL 1 is less than the purchase price at CSL 0. Consumers are said to have no impact if the
base case forecast product assigned to them has a greater efficiency than the level indicated by a
standard.
Table 8.4.14 LCC and PBP Results for Low Energy, AC Out Battery Chargers
(Test/Teardown Data)

CSL
0
1

UEC
kWh
19.6
6.4

Avg.
Installed
Price
2009$
5.93
4.54

Life-Cycle Cost
Avg.
Operating
Cost
2009$
11.07
3.60

Avg.
LCC
2009$
17.00
8.14

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings
Cost
Impact
Benefit
2009$
%
%
%
7.88
0.0
11.1
88.9

Payback Period

Median
yrs.
0.0

Mean
yrs.
0.0

8.4.4 Battery Charger Results (Scaled Product Classes)


DOE conducted a life-cycle cost analysis on four scaled BC product classes at different
combinations of battery energy and battery voltage, one for each of the four scaled BC product
classes: low energy, medium voltage; medium energy, high voltage; DC-DC, <9V input; and
DC-DC, 9V input. As discussed in chapter 5, the cost-efficiency curves for each of the four
scaled BC product classes were derived by scaling data from the representative units. DOE only
considered manufacturer-provided data when scaling to these product classes.
Even though the scaled BC product classes are derived from the representative product
classes, they still exhibit different LCC results. In general, the scaled BC product classes
resulted in lower LCC savings than their corresponding representative product classes. This is
attributable to three main factors. First, many of the scaled product classes have a lighter usage
profile than the corresponding representative product class, resulting in less opportunity for
operating cost savings. Second, several of the scaled product classes have lower battery energies
than their corresponding representative product class, which leads to lower total losses, and thus
smaller incremental savings from one CSL to another. Finally, some scaled product classes lack
an EPS whereas their corresponding representative product classes do not. As a result, they do
not suffer the losses associated with an EPS, and thus have less opportunity for incremental
operating cost savings.

8-63

Similar to the representative BC units discussed previously, the MSPs for each efficiency
level considered in the BC analysis represent the total MSP attributable to a product, not the
incremental MSP over a baseline level. As such, the average installed prices can be interpreted
as the total installed cost at each CSL, and the total life-cycle cost is representative of the actual
cost associated with the product. The LCC savings, PBP calculations, and percentage of
consumers with a net cost or net benefit are unaffected by the type of MSP used, and thus remain
accurate.
8.4.4.1 Base Case LCC Distributions (Manufacturer Data)
Figure 8.47 through Figure 8.50 show the frequency charts for the base case LCC for the
four scaled BC product classes using manufacturer data. Since the base case considers the
different efficiency levels at which consumers currently purchase products, the base case LCC
distribution is composed of several standards-levels. If all consumers purchased products at CSL
0 in the base case, then the LCC distribution would be composed entirely of the LCC at CSL 0.
However, the LCC distribution in the base case shows the LCC results in proportion to how
many users currently purchase products at each standards-level (CSLs 0 through 3).
To find the appropriate distribution of LCCs in the base case, the total LCC was
calculated for each standards-level considered (CSLs 0 through 3). Then these distributions were
weighted by the market shares for each standards-level in the base case, as discussed in 8.2.10.
The combined chart shows the aggregated LCC distribution across all CSLs in the base case.

14%
12%
10%
8%
CSL 3

6%

CSL 2

4%

CSL 1

2%

CSL 0

0%
6.50
7.80
9.10
10.40
11.70
13.00
14.30
15.60
16.90
18.20
19.50
20.80
22.10
23.40
24.70
26.00
27.30
28.60
29.90
31.20
32.50

Percent of Consumers (%)

Total LCC in the Base Case


(weighted by market distribution)

Weighted Total LCC ($)

Figure 8.47 Low Energy, Medium Voltage Battery Charger: Base Case LCC Distribution
(Manufacturer Data)

8-64

7%
6%
5%
4%
3%
2%
1%
0%

CSL 3
CSL 2

CSL 1
CSL 0

61.10
69.10
77.20
85.20
93.20
101.30
109.30
117.30
125.30
133.40
141.40
149.40
157.50
165.50
173.50
181.50
189.60
197.60
205.60
213.70
221.70

Percent of Consumers (%)

Total LCC in the Base Case


(weighted by market distribution)

Weighted Total LCC ($)

Figure 8.48 Medium Energy, High Voltage Battery Charger: Base Case LCC Distribution
(Manufacturer Data)

30%
25%
20%
15%

CSL 3

10%

CSL 2

CSL 1

5%

CSL 0

0%
1.30
1.50
1.70
1.80
2.00
2.10
2.30
2.50
2.60
2.80
3.00
3.10
3.30
3.50
3.60
3.80
4.00
4.10

Percent of Consumers (%)

Total LCC in the Base Case


(weighted by market distribution)

Weighted Total LCC ($)

Figure 8.49 DC-DC, <9V Input Battery Charger: Base Case LCC Distribution
(Manufacturer Data)

8-65

35%
30%

25%
20%

CSL 3

15%

CSL 2

10%

CSL 1

5%

CSL 0

7.50

7.20

6.90

6.60

6.30

6.00

5.70

5.40

5.10

4.80

4.50

4.20

3.90

3.60

0%
3.30

Percent of Consumers (%)

Total LCC in the Base Case


(weighted by market distribution)

Weighted Total LCC ($)

Figure 8.50 DC-DC, 9V Input Battery Charger: Base Case LCC Distribution
(Manufacturer Data)
8.4.4.2 Summarized Life-Cycle Cost and Payback Period Results
Table 8.4.15 through Table 8.4.18 summarizes the LCC and PBP results for the four
scaled BC product classes using manufacturer data. As mentioned earlier, for some consumers,
DOE assigned products in the base case that are more energy efficient than some of the energy
levels under consideration. For that reason, the average LCC impacts are not equal to the
difference between the LCC of a specific energy efficiency level and the LCC of the baseline
product. Similarly with regard to PBPs shown below, DOE determined the median and average
values by excluding the percentage of consumers not impacted by a standard at a given
efficiency level. The values for average operating cost in the table are discounted sums of the
annual operating costs over the product lifetime.
Table 8.4.15 LCC and PBP Results for Low Energy, Medium Voltage Battery Chargers
(Manufacturer Data)

CSL
0
1
2

UEC
kWh
10.0
5.4
1.0

Avg.
Installed
Price
2009$
7.38
9.94
21.42

Life-Cycle Cost
Avg.
Operating
Cost
2009$
4.88
2.63
0.45

Avg.
LCC
2009$
12.25
12.57
21.87

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings
Cost
Impact
Benefit
2009$
%
%
%
-0.07
25.6
66.0
8.4
-8.23
83.4
12.0
4.6

8-66

Payback Period

Median
yrs.
7.5
32.6

Mean
yrs.
17.0
88.3

Table 8.4.16 LCC and PBP Results for Medium Energy, High Voltage Battery Chargers
(Manufacturer Data)

CSL
0
1
2
3

UEC
kWh
69.3
41.9
25.0
18.0

Avg.
Installed
Price
2009$
45.14
70.52
130.31
222.45

Life-Cycle Cost
Avg.
Operating
Cost
2009$
54.91
34.17
20.37
14.75

Avg.
LCC
2009$
100.06
104.69
150.68
237.20

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings
Cost
Impact
Benefit
2009$
%
%
%
-2.43
41.5
50.0
8.5
-34.75
69.9
30.0
0.1
-121.27 100.0
0.0
0.0

Payback Period

Median
yrs.
8.8
20.2
41.5

Mean
yrs.
8.9
28.1
74.6

Table 8.4.17 LCC and PBP Results from DC-DC, <9V Input Battery Chargers
(Manufacturer Data)

CSL
0
1
2

UEC
kWh
1.5
0.9
0.8

Avg.
Installed
Price
2009$
1.41
2.59
3.69

Life-Cycle Cost
Avg.
Operating
Cost
2009$
0.46
0.27
0.17

Avg.
LCC
2009$
1.87
2.86
3.86

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings
Cost
Impact
Benefit
2009$
%
%
%
-0.39
39.2
60.8
0.0
-1.36
97.2
2.8
0.0

Payback Period

Median
yrs.
17.7
26.3

Mean
yrs.
23.1
40.2

Table 8.4.18 LCC and PBP Results from DC-DC, 9V Input Battery Chargers
(Manufacturer Data)

CSL
0
1
2

8.5

UEC
kWh
1.3
0.8
0.5

Avg.
Installed
Price
2009$
1.73
2.61
5.60

Life-Cycle Cost
Avg.
Operating
Cost
2009$
2.36
1.50
0.89

Avg.
LCC
2009$
4.09
4.11
6.48

Life-Cycle Cost Savings


Consumers with
Wtd.
Avg.
Net
No
Net
Savings
Cost
Impact
Benefit
2009$
%
%
%
-0.01
30.1
61.1
8.8
-2.32
96.5
2.8
0.7

Payback Period

Median
yrs.
20.4
75.2

Mean
yrs.
19.0
72.7

DETAILED RESULTS AND SENSITIVITY RUNS

DOE presents more detailed results and sensitivity runs from the LCC analysis in
Appendix 8B. These results include electricity price sensitivity scenarios, commercial sector
sensitivity scenarios, and Monte Carlo probability distributions for high-savings and low-savings
scenarios.

8-67

REFERENCES

U.S. Department of Energy. Energy Information Administration. Annual Energy Outlook 2010

Early Release. March, 2010. Washington, D.C.

http://www.eia.doe.gov/oiaf/aeo/.

ii

U.S. Department of Energy. Energy Information Administration. Energy Market and

Economic Impacts of S.2191, The Lieberman-Warner Climate Security Act of 2007. April 2008.

Washington, D.C.

http://www.eia.doe.gov/oiaf/servicerpt/s2191/excel/s2191.xls.

iii

U.S. Department of Energy. Energy Information Administration. Form EIA-861 Final Data
File for 2008. March, 2010. Washington, D.C.
http://www.eia.doe.gov/cneaf/electricity/page/eia861.html.
iv

U.S. Department of Energy. Energy Information Administration. Annual Energy Outlook 2009
with Projections to 2030. March, 2009. Washington, D.C.
http://www.eia.doe.gov/oiaf/aeo/index.html.

Title 10, Code of Federal Regulations, Part 474, Subpart C Electric and Hybrid Vehicle
Research, Development, and Demonstration Program. June 12, 2000.
http://www.gpo.gov/fdsys/pkg/FR-2000-06-12/html/00-14446.htm.

vi

Bradfield, M. Improving Alternator Efficiency Measurably Reduces Fuel Costs. 2008. Remy

International, Inc: Pendleton, IN.

vii

The Federal Reserve Board, Survey of Consumer Finances 1989, 1992, 1995, 1998, 2001,

2004, 2007.

http://www.federalreserve.gov/pubs/oss/oss2/scfindex.html.

viii

The Federal Reserve Board, Federal Reserve Statistical Release, Selected Interest Rates,

Historical Data, Instrument: CDs (secondary market), Maturity: 6-month, Frequency: Annual,

Description: Average rate on 6-month negotiable certificates of deposit (secondary market),

quoted on an investment basis, 2010.

http://www.federalreserve.gov/releases/H15/data.htm.

ix

The Federal Reserve Board, Federal Reserve Statistical Release, Selected Interest Rates,
Historical Data, Instrument: State and local bonds, Maturity: 20-year, Frequency: Monthly,
Description: Bond buyer go 20-bond municipal bond index, 2010.
http://www.federalreserve.gov/releases/H15/data.htm.

The Federal Reserve Board, Federal Reserve Statistical Release, Selected Interest Rates,
Historical Data, Instrument: Corporate bonds/Moody's Seasoned AAA, Frequency: Annual,
Description: Moody's yield on seasoned corporate bonds - all industries, AAA, 2010.
http://www.federalreserve.gov/releases/H15/data.htm.

8-68

xi

Mortgage-X - Mortgage Information Service, Cost of Savings Index (COSI) Index History,

2010.

http://mortgage-x.com/general/indexes/default.asp.

xii

Damodaran Online Data Page, Historical Returns on Stocks, Bonds and Bills-United States,

2010. Damodaran.

http://pages.stern.nyu.edu/~adamodar.

xiii

Modigliani, F. and M. Miller. The Cost of Capital, Corporation Finance and the Theory of
Investment. American Economic Review. 1958. (June): pp. 261-297.
xiv

Ibbotsons Associates. Stocks, Bonds, Bills, and Inflation, Valuation Edition, 2001 Yearbook.
2001.

xv

Damodaran, A. Damodaran Online. Current US data.


http://pages.stern.nyu.edu/~adamodar.

xvi

Executive Office of the President. 2010 Economic Report of the President, U.S. Government
Printing Office, Washington, D.C., p. 75.
http://www.whitehouse.gov/sites/default/files/microsites/economic-report-president.pdf.

xvii

Value Line. Value Line Investment Survey. 2010.


http://www.valueline.com.
xviii

U.S. Office of Management and Budget. Circular No. A-94. Appendix C. 2009.
http://www.whitehouse.gov/omb/circulars_a094_a94_appx-c/.
xix

U.S. Census Bureau. The 2010 Statistical Abstract. Table 607 Employment by Industry.
http://www.census.gov/compendia/statab/2010/tables/10s0607.xls.
xx

U.S. Census Bureau. The 2010 Statistical Abstract. Table 484 Federal Civilian
Employment and Annual Payroll by Branch.
http://www.census.gov/compendia/statab/2010/tables/10s0484.xls.
xxi

U.S. Census Bureau. Government Employment and Payroll. 2008 State and Local

Government.

http://www2.census.gov/govs/apes/08stlall.xls.

8-69

CHAPTER 9. SHIPMENTS ANALYSIS


TABLE OF CONTENTS
9.1
9.2

9.3

9.4

INTRODUCTION ........................................................................................................... 9-1

BC AND EPS SHIPMENTS IN THE BASE CASE ....................................................... 9-1

9.2.1 Shipments Forecast .............................................................................................. 9-1

9.2.2 Efficiency Forecast .............................................................................................. 9-2

EFFECT OF STANDARDS ON BC AND EPS SHIPMENTS ...................................... 9-3

9.3.1 Efficiency of BCs and EPSs ................................................................................ 9-3

9.3.2 Price Elasticity of Demand for BCs and EPSs..................................................... 9-4

9.3.3 Substitution Away from BCs and EPSs............................................................... 9-4

RESULTS ........................................................................................................................ 9-6

9.4.1 External Power Supply Shipments Forecast ........................................................ 9-6

9.4.2 Battery Charger Shipments Forecast ................................................................... 9-7

LIST OF TABLES

TABLE 9.1 ROLL-UP MARKET RESPONSE FOR A HYPOTHETICAL BC OR EPS

PRODUCT CLASS ..................................................................................................................... 9-4

TABLE 9.2 EXTERNAL POWER SUPPLY SHIPMENTS BY PRODUCT CLASS............... 9-6

TABLE 9.3 BASE CASE EXTERNAL POWER SUPPLY EFFICIENCY IN 2013................. 9-7

TABLE 9.4 BATTERY CHARGER SHIPMENTS BY PRODUCT CLASS ............................ 9-8

TABLE 9.5 BASE CASE BATTERY CHARGER EFFICIENCY IN 2013 .............................. 9-9

9-i

CHAPTER 9. SHIPMENTS ANALYSIS


9.1

Introduction

This chapter describes the data and methods that DOE used to generate shipment
forecasts and base case efficiency distributions for each of the product classes being considered
in this preliminary analysis of standards for battery chargers and external power supplies.
Outputs from the shipments analysis are inputs to the life-cycle cost analysis (chapter 8), national
impact analysis (chapter 10), and manufacturer impact analysis (chapter 12).
The calculations for shipment forecasts were implemented as part of the National Impact
Analysis (NIA). These calculations are contained in the NIA Microsoft Excel workbook that
can be downloaded from the EERE web site.1 The workbook, entitled BCEPS Preliminary
Analysis - NIA.xls contains a tab for each of the product classes analyzed as part of the
rulemaking.
This document is a guide to the inputs and methodology employed in the workbook.
Section 9.2 presents the methodology for developing a base case shipments forecast. Section 9.3
discusses the potential impacts of standards on the shipments forecast. The outputs from the
shipments analysis are shown in Section 9.4.
9.2

BC and EPS Shipments in the Base Case

The shipment analysis consists of two outputs:


A shipments forecast, which calculates the total number of BCs and EPSs shipped each
year over a 30 year period, beginning in 2013 and ending in 2042.
An efficiency forecast, which shows the distribution of shipments of BCs and EPSs by
candidate standard level (CSL), which determines the percentage of shipments affected
by a standard.
9.2.1 Shipments Forecast
To develop its shipments forecast, DOE combined current year (2008) shipments,
discussed in the market assessment (chapter 3) with a compound annual growth rate for BCs and
EPSs, discussed in this section. Shipment values were calculated for 30 years, through 2042, the
last year of the analysis period. Shipments for 2008, 2013 and 2042 are presented in Table 9.2
(EPSs) and Table 9.4 (BCs).
As discussed in the market assessment (chapter 3), the variety of product applications that
employ BCs and EPSs is vast. DOE is aware that this mix of product applications will very likely
change dramatically over the analysis period. Indeed, most of the end-use products that now
employ BCs or EPSs were not on the market twenty years ago. Therefore, forecasting the size of
the market more than twenty years into the future is extremely difficult.
1

Available for download from the EERE website:


http://www.eere.energy.gov/buildings/appliance_standards/residential/battery_external.html
9-1

In its research regarding the market for BCs and EPSs, DOE noted that the market for
these products has grown tremendously in the past 10 years. Additionally, DOE found that many
market reports have predicted enormous future growth for the applications that employ BCs and
EPSs. However, in forecasting the size of the BC and EPS markets over the next 32 years, DOE
considered the possibility that much of the market growth associated with these products has
already occurred. For many reports predicting growth of applications that employ BCs or EPSs,
DOE noted that growth was predicted for new applications, but older applications were generally
not included. That is, the demand for BCs and EPSs had not grown, but rather the products that
use BCs and EPSs had transitioned to a new product mix.
With this in mind, in its forecast DOE took a conservative approach and assumed that
while the specific applications that use BCs or EPSs will change, the overall number of
individual units that use BCs or EPSs will grow slowly, with new applications replacing some
current applications, but with little change in per-capita ownership of BCs or EPSs over time.
To estimate future market size while assuming no change in the per-capita BC and EPS
purchase rate, DOE used population growth rate as the compound annual market growth rate.
DOE feels this growth rate represents a conservative approximation of the expected market
progression for these products. Population growth rate values were obtained from the U.S.
Census Bureau 2009 National Projections, which forecast population through 2050. DOE took
the average annual population growth rate, 0.75 percent, and applied this rate to all BC and EPS
product classes.
9.2.2 Efficiency Forecast
To evaluate the potential impacts of standards, DOE developed a base case efficiency
forecast, which represents DOEs estimate of the future state of the market with respect to
efficiency if energy conservation standards for the units covered under this rulemaking are not
adopted (or, in the case of Class A EPSs, strengthened). The impact of a standard is then the
relative improvement in efficiency compared to this forecast.
DOEs starting point in developing base case efficiency forecasts was current year
efficiency distributions, as described in the market assessment (chapter 3). To extrapolate from
the present day forward to 2013, the first year of the analysis period, DOE looked at recent
trends in product efficiency and considered what factors might lead BCs and EPSs to become
more efficient between now and 2013.
For BCs, DOE found that with the exception of Canada, there were no BC standards
slated to take effect by 2013. In addition, after reviewing the ENERGY STAR voluntary
program for battery charging systems, DOE found that as of January 22, 2010, less than 150
battery charging systems had been qualified.2 Therefore, DOE could find no compelling
evidence that battery charging systems will improve in efficiency in the short-term.

http://www.energystar.gov/index.cfm?c=battery_chargers.pr_battery_chargers
9-2

For EPSs, DOE found two programs that would influence EPS efficiency in the shortterm. The first is the ENERGY STAR program for EPSs (called external power adapters),
which currently requires that EPSs be at or above CSL 1 in order to qualify. This voluntary
program is very active, with several thousand qualified products as of January 22, 2010.3 The
second are the European Union Ecodesign requirements on Energy Using Products, which
include legislation on EPSs that requires that EPSs sold in the E.U. be at or above CSL 1,
effective April 2011. Europe currently represents approximately one-third of the global EPS
market. DOE could find no future program which required efficiency above CSL 1. These
factors apply to Class A EPSs. With this in mind, DOE estimated that approximately half of the
Class A EPS market at CSL 0 in 2008 would transition to CSL 1 by 2013. DOE plans to update
this estimate in the NOPR stage of the analysis.
In addition, DOE expects that efficiency improvements to Class A EPSs are likely to spill
over into product classes M1, M2, M3, M4, B1, B2, B3 and B4 because the construction of these
non-Class A EPSs is very similar to the construction of Class A EPSs. Therefore, DOE assumes
that EPSs in these product classes will exhibit the same efficiency improvements as Class A
EPSs between now and 2013. Many of the EPSs in product class X2 are already at CSL 1; DOE
therefore assumes that there will be little or no further improvements in the efficiency of these
products in the short term. DOE also assumes no short-term improvement in the efficiency of
products in product classes X1 and H1, as there are no current or planned standards programs
that are expected to affect these EPSs.
In light of the above considerations, DOE developed base case efficiency forecasts in
which there are modest improvements in EPS efficiency and no improvement in BC efficiency
between now and 2013. These efficiency forecasts are shown in Table 9.3 for EPSs and Table
9.5 for BCs. DOE believes that these forecasts provide a reasonable reference point for assessing
the impact of potential standards.
DOE next evaluated the likelihood that BCs and EPSs would continue to improve in
efficiency throughout the analysis period. However, DOE found that although efficiencies might
improve between now and 2013, no data exist to suggest that BC or EPS efficiencies would
improve further during the following 30 years in the absence of standards. Therefore, in the
preliminary analysis DOE forecasts static efficiency distributions for both BCs and EPSs
throughout the analysis period.
9.3

Effect of Standards on BC and EPS Shipments

9.3.1 Efficiency of BCs and EPSs


In addition to quantifying the projected impact of standards on total shipments, DOE
must also consider the change in the mix of product efficiencies due to standards. In its
preliminary analysis, DOE assumed that manufacturers will respond to standards by improving
those products that do not meet the standards to the standard level, but no higher, while the
products that were already as or more efficient than the standard remain unaffected. This is
referred to as a roll-up response to standards.
3

http://www.energystar.gov/index.cfm?c=ext_power_supplies.power_supplies_consumers
9-3

The mechanics of a roll-up response are detailed in Table 9.1. The Base Case gives the
efficiency distribution with no standard. In the Standard Set at CSL 1 scenario, all the
shipments from CSL 0 are rolled up to CSL 1, the level of the standard. The same methodology
is applied to the other standards cases.
Table 9.1 Roll-Up Market Response for a Hypothetical BC or EPS Product Class
Case
Base Case
Standard Set at CSL 1
Standard Set at CSL 2
Standard Set at CSL 3
Standard Set at CSL 4

CSL 0
25%
0%
0%
0%
0%

Percent of Market at Each Efficiency Level


CSL 1
CSL 2
CSL 3
50%
25%
0%
75%
25%
0%
0%
100%
0%
0%
0%
100%
0%
0%
0%

CSL 4
0%
0%
0%
0%
100%

9.3.2 Price Elasticity of Demand for BCs and EPSs


In other rulemakings, forecasted shipments in the standards case typically deviate from
the base case. The magnitude of the difference between the standards case and base case
shipments forecasts depends on the calculated purchase price increase and the operating cost
savings from the standard. Because the purchase price tends to have a larger impact than
operating cost on equipment purchase decisions, standards case forecasts typically show
elasticity of demand, manifested as a change (often a decrease) in shipments relative to the base
case. However, in the case of BCs and EPSs, DOE expects demand to be fairly inelastic, as the
incremental cost of more efficient BCs and EPSs is, in most cases, small relative to the total cost
of the end-use product and is hidden from the consumer as just a portion of the total cost of the
end-use product. The incremental final product price of Class A EPSs at CSL 4, for example,
ranges between $0.54 and $6.31.
Thus, in its preliminary analysis, DOE makes the simplifying assumption that the demand
for BCs and EPSs is perfectly inelastic; that is, a price increase in the standards case will not lead
to a decrease in demand for the product.
9.3.3 Substitution Away from BCs and EPSs
Another potential market response to the presence of a standard for BCs or EPSs would
be to substitute a different power source for the BC or EPS. This could reduce BC or EPS
shipments in the standards case relative to the base case. However, for both BCs and EPSs, the
extent to which manufacturers choose substitute power sources will be limited by design
constraints.
Possible substitutions for BCs include mains power. DOE considers the possibility of
substitution to be minimal, since such a substitution would remove the primary functionality that
battery chargers offer, portability.

9-4

Possible substitutions for EPSs include internal power supplies, batteries, or USB power.
As with BCs, DOE considers the potential for substitution to be minimal. In most cases, the
choice of an external power supply over an internal power supply is to minimize the size of the
application. Use of batteries would not eliminate the usage of an EPS. The last substitute, USB
power, may have some impact; however, DOE estimates this impact to be limited because USB
ports are much less common than traditional wall outlets.
Thus, in its preliminary analysis, DOE assumes that the impact of substitution for BCs
and EPSs is negligible and, thus, does not attempt to quantify it in the standards case.

9-5

9.4

Results

9.4.1 External Power Supply Shipments Forecast


In DOEs forecast, Class A EPS shipments grow from 289 million in 2008 to 372 million
in 2042. Table 9.2 shows DOEs shipments forecast for each of the four Class A EPS product
classes, as well as the eleven non-Class A EPS product classes.
Table 9.2 External Power Supply Shipments by Product Class
Class
ID

2008
Shipments
(estimated)

2013
Shipments
(projected)

2042
Shipments
(projected)

Basic Voltage

A1

206,176,000

214,025,000

265,810,000

Low Voltage

A2

72,195,000

74,943,000

93,076,000

Basic Voltage

A3

7,994,000

8,298,000

10,306,000

Low Voltage

A4

2,250,000

2,336,000

2,901,000

100 W

X1

3,782,000

3,926,000

4,876,000

100 W

X2

4,901,000

5,088,000

6,319,000

High Power

>250 W

H1

3,000

3,000

4,000

AC
For Medical DC
Devices
AC

Basic Voltage

M1

1,534,000

1,592,000

1,978,000

Low Voltage

M2

25,000

26,000

32,000

Basic Voltage

M3

AC

Low Voltage

M4

AC
DC

Basic Voltage

B1

1,884,000

1,955,718

2,429,000

Low Voltage

B2

275,000

285,000

355,000

Basic Voltage

B3

Low Voltage

B4

Output
Class A
AC-DC
AC-AC
Non-Class A
Multiple-Voltage

For MotorOperated
BC
Applications
and
Detachable
Batteries

AC
AC

Table 9.3 shows DOEs assumptions about the efficiency of EPSs in 2013 in the base
case. The percentages show, for each product class, what fraction of new products sold each year
are at each efficiency level (CSL). These market shares are assumed to remain constant
throughout the analysis period, which begins in 2013.

9-6

Table 9.3 Base Case External Power Supply Efficiency in 2013


Output

Class
ID

Percentage of Market at Each CSL


CSL 0

CSL 1 CSL 2 CSL 3 CSL 4

CSL 5

Class A

AC-DC

AC-AC

Basic
Voltage

A1

34

51

12

Low
Voltage

A2

34

51

12

Basic
Voltage

A3

34

51

12

Low
Voltage

A4

34

51

12

100 W

X1

100

100 W

X2

95

>250 W
Basic
Voltage
Low
Voltage
Basic
Voltage
Low
Voltage
Basic
Voltage
Low
Voltage
Basic
Voltage
Low
Voltage

H1

60

40

M1

10

29

46

12

M2

10

29

46

12

M3

10

29

46

12

M4

10

29

46

12

B1

10

29

46

12

B2

10

29

46

12

B3

10

29

46

12

B4

10

29

46

12

Non-Class A
Multiple-Voltage
High Power
AC
DC
For Medical
Devices
AC
AC
For MotorOperated BC AC
Applications DC
and
Detachable
Batteries
AC
AC

9.4.2 Battery Charger Shipments Forecast


In DOEs forecast, BC shipments grow from 402 million in 2008 to 519 million in 2042.
Table 9.4 shows DOEs shipments forecast for each of the 10 BC product classes.

9-7

Table 9.4 Battery Charger Shipments by Product Class


Battery
Energy

Battery
Voltage
Inductive
Connection

AC
DC

<100 Wh <4 V
4<10 V
100
3000 Wh

AC-AC

1
2

2008
Shipments
(estimated)
5,354,000

2013
Shipments
(projected)

2042
Shipments
(projected)

5,558,000

6,903,000

206,221,000 214,071,000 265,867,000

23,116,000

23,996,000

29,802,000

10 V

69,758,000

72,413,000

89,934,000

<20 V

2,333,000

2,422,000

3,007,000

20 V

940,000

976,000

1,212,000

214,000

222,000

276,000

<9 V Input

71,825,000

74,560,000

92,600,000

9 V Input

15,519,000

16,110,000

20,008,000

AC Output
from
Battery

10

6,900,000

7,163,000

8,896,000

>3000 Wh
DC-DC

Class
ID

9-8

Table 9.5 shows DOEs assumptions about the efficiency of BCs in 2013 in the base
case. The percentages show, for each product class, what fraction of new products sold each year
are at each efficiency level (CSL). These market shares are assumed to remain constant
throughout the analysis period, which begins in 2013.
Table 9.5 Base Case Battery Charger Efficiency in 2013
Battery
Energy

<100
Wh
AC
DC
100
3000
Wh

Battery
Voltage

CSL 0

CSL 1

CSL 2

CSL 3

Inductive
Connection

75

12.5

12.5

<4 V

39

58

4<10 V

34

54

12

10 V

19

43

38

<20 V

43

15

40

20 V

50

20

30

50

20

30

<9 V Input

39

58

9 V Input

39

58

AC Output
from
Battery

10

67

22

11

>3000 Wh
DC-DC

AC-AC

Percent of Market at Each CSL

Class
ID

9-9

CHAPTER 10 NATIONAL IMPACT ANALYSIS

TABLE OF CONTENTS
10.1 Introduction................................................................................................................. 10-1

10.1.1 National Inventory Accounting Model .............................................................. 10-4

10.1.2
Calculation of Product Class Lifetime Profiles ................................................. 10-4

10.1.3
National Inventory Example .............................................................................. 10-5

10.2 National Energy Savings............................................................................................. 10-7

10.2.1
Calculating National Energy Savings ................................................................ 10-7

10.2.2
Inputs to National Energy Savings Calculation ................................................. 10-7

10.2.2.1 Unit Energy Savings ........................................................................................ 10-8

10.2.2.2 Site-to-Source Conversion Factors ................................................................ 10-10

10.3 Net Present Value ..................................................................................................... 10-11

10.3.1
Calculating Net Present Value ......................................................................... 10-11

10.3.2 Inputs to Net Present Value Calculation.......................................................... 10-12

10.3.2.1 Electricity Prices ............................................................................................ 10-12

10.3.2.2 Gasoline Prices............................................................................................... 10-12

10.3.2.3 Unit Improvement Costs ................................................................................ 10-12

10.3.2.4 Discount Rates ............................................................................................... 10-14

10.4 Results....................................................................................................................... 10-14

10.4.1
National Energy Savings.................................................................................. 10-15

10.4.2
Annual Costs and Savings ............................................................................... 10-17

10.4.3 Net Present Value ............................................................................................ 10-18

LIST OF TABLES
TABLE 10.1 NATIONAL ANNUAL ENERGY CONSUMPTION OF EXTERNAL POWER

SUPPLIES SHIPPED IN 2008............................................................................... 10-1

TABLE 10.2 NATIONAL ANNUAL ENERGY CONSUMPTION OF BATTERY CHARGERS

SHIPPED IN 2008 .................................................................................................. 10-2

TABLE 10.3 EXAMPLE PRODUCT CLASS CHARACTERISTICS FOR CALCULATION OF

LIFETIME PROFILE ............................................................................................. 10-4

TABLE 10.4 EXAMPLE LIFETIME PROFILE ...................................................................... 10-4

TABLE 10.5 EXAMPLE NATIONAL INVENTORY TABLE ............................................... 10-5

TABLE 10.6. SHIPMENT-WEIGHTED AVERAGE UNIT ENERGY SAVINGS FOR

EXTERNAL POWER SUPPLIES (KWH/YR) ..................................................... 10-8

TABLE 10.7. SHIPMENT-WEIGHTED AVERAGE UNIT ENERGY SAVINGS FOR

BATTERY CHARGERS, MANUFACTURER DATA (KWH/YR) .................... 10-9

TABLE 10.8. SHIPMENT-WEIGHTED AVERAGE UNIT ENERGY SAVINGS FOR

BATTERY CHARGERS, TEST AND TEARDOWN DATA (KWH/YR) .......... 10-9

10-i

TABLE 10.9 SHIPMENT-WEIGHTED AVERAGE UNIT IMPROVEMENT COSTS FOR


EXTERNAL POWER SUPPLIES (2009$) MANUFACTURER DATA ........ 10-13
TABLE 10.10 SHIPMENT-WEIGHTED AVERAGE UNIT IMPROVEMENT COSTS FOR
BATTERY CHARGERS (2009$) MANUFACTURER DATA....................... 10-13
TABLE 10.11 SHIPMENT-WEIGHTED AVERAGE UNIT IMPROVEMENT COSTS FOR
BATTERY CHARGERS (2009$) TEST AND TEARDOWN DATA............. 10-14
TABLE 10.12 NATIONAL ENERGY SAVINGS POTENTIAL FROM AMENDED
STANDARDS FOR EXTERNAL POWER SUPPLIES (TRILLION BTU) ...... 10-15
TABLE 10.13 NATIONAL ENERGY SAVINGS POTENTIAL FROM NEW STANDARDS
FOR BATTERY CHARGERS, MANUFACTURER DATA (TRILLION BTU)10-16
TABLE 10.14 NATIONAL ENERGY SAVINGS POTENTIAL FROM NEW STANDARDS
FOR BATTERY CHARGERS, TEST AND TEARDOWN DATA (TRILLION
BTU) ..................................................................................................................... 10-17
TABLE 10.15 NET PRESENT VALUE OF CONSUMER SAVINGS FROM AMENDED
STANDARDS FOR EXTERNAL POWER SUPPLIES AT A 3 PERCENT
DISCOUNT RATE, MANUFACTURER DATA (2009$ MILLIONS).............. 10-19
TABLE 10.16 NET PRESENT VALUE OF CONSUMER SAVINGS FROM AMENDED
STANDARDS FOR EXTERNAL POWER SUPPLIES AT A 7 PERCENT
DISCOUNT RATE, MANUFACTURER DATA (2009$ MILLIONS).............. 10-19
TABLE 10.17 NET PRESENT VALUE OF CONSUMER SAVINGS FROM NEW
STANDARDS FOR BATTERY CHARGERS AT A 3 PERCENT DISCOUNT
RATE, MANUFACTURER DATA (2009$ MILLIONS) ................................... 10-20
TABLE 10.18 NET PRESENT VALUE OF CONSUMER SAVINGS FROM NEW
STANDARDS FOR BATTERY CHARGERS AT A 7 PERCENT DISCOUNT
RATE, MANUFACTURER DATA (2009$ MILLIONS) ................................... 10-21
TABLE 10.19 NET PRESENT VALUE OF CONSUMER SAVINGS FROM NEW
STANDARDS FOR BATTERY CHARGERS AT A 3 PERCENT DISCOUNT
RATE, TEST AND TEARDOWN DATA (2009$ MILLIONS)......................... 10-22
TABLE 10.20 NET PRESENT VALUE OF CONSUMER SAVINGS FROM NEW
STANDARDS FOR BATTERY CHARGERS AT A 7 PERCENT DISCOUNT
RATE, TEST AND TEARDOWN DATA (2009$ MILLIONS) ......................... 10-23

LIST OF FIGURES
FIGURE 10.1 FLOW CHART SHOWING THE CALCULATION OF NATIONAL ENERGY
SAVINGS AND NET PRESENT VALUE............................................................ 10-3
FIGURE 10.2 SITE-TO-SOURCE CONVERSION FACTORS, 2010 TO 2060................... 10-10

FIGURE 10.3 REFERENCE CASE ANNUAL UNDISCOUNTED CONSUMER COSTS AND


SAVINGS FROM A STANDARD AT CSL 1 FOR AC-DC BASIC VOLTAGE
CLASS A EPSS WITH NAMEPLATE OUTPUT POWER UP TO 10.25 WATTS10-18

10-ii

CHAPTER 10 NATIONAL IMPACT ANALYSIS


10.1 Introduction
This chapter describes the methods the U.S. Department of Energy (DOE) used to
estimate the national consumer impacts of standards for BCs and EPSs. DOE evaluated the
following impacts: (1) national energy savings (NES) attributable to each possible standard, (2)
monetary value of those energy savings to consumers of the considered products, (3) increased
cost of the products because of standards, and (4) net present value (NPV) of total consumer
costs and savings (the value of energy savings less increased product costs).
For reference, if current EPS and BC average efficiencies were applied to shipments from
2008, those units would consume approximately 2.2 billion kWh and 3.4 billion kWh of site
electricity, respectively, or 24 trillion Btu and 37 trillion Btu of primary energy, respectively,
each year, as detailed in Table 10.1 and Table 10.2. Values in these tables reflect the inputs used
in the reference savings case, as described in appendix 8B.
Table 10.1 National Annual Energy Consumption of External Power Supplies Shipped in
2008
Annual Energy Consumption
Product Class

Site Energy
(Million kWh)

Primary Energy Percent of Total


(Trillion Btu)

A1: 0-10.25 W

251

11%

A1: 10.25-39 W

930

10

42%

A1: 39-90 W

582

26%

A1: 90-250 W

186

8%

A2 (DC Output, Low Voltage)

144

6%

A3 (AC Output, Basic Voltage)

113

5%

A4 (AC Output, Low Voltage)

17

1%

2,222

24

100%

All Class A EPSs

Note: Figures calculated using DOEs estimates of the efficiency of products in the market in 2009.
Values may not sum to totals due to rounding.

10-1

Table 10.2 National Annual Energy Consumption of Battery Chargers Shipped in 2008
Annual Energy Consumption
Conversion
Type

Battery
Energy

<100 Wh
AC-DC
100
3000 Wh

Battery
Voltage

AC-AC

Site Energy
(Million kWh)

Primary
Energy
(Trillion Btu)

47

1%

Inductive
Connection

<4 V

1,539

17

45%

4<10 V

148

4%

10 V

991

11

29%

<20 V

359

11%

20 V

48

1%

55

2%

<9 V Input

80

2%

9 V Input

15

0.3

<1%

AC Output
from
Battery

10
125

4%

3,408

37

100%

>3000 Wh
DC-DC

Class
ID

All Battery Chargers

DOE determined both the NES and NPV for each candidate standard level (CSLs) for
BCs and EPSs. These calculations are described in detail as follows:
National Inventory, discussed in section 10.1.1, details the methodology for
calculating the number of units in use in a given year that are subject to a standard.
NES, discussed in section 10.2, combines the average change in unit energy
consumption (UEC) due to a standard (unit energy savings) with the national
inventory to obtain the aggregate energy savings generated by a standard.
NPV, discussed in Section 10.3, compares the present value (in 2009$) of the NES
with the present value (in 2009$) of the national improvement costs associated with a
standard. National improvement cost combines the average change in unit cost (unit
improvement cost) with the number of units shipped in a given year to obtain the
aggregate incremental costs paid by consumers due to a standard.
The national impact analysis (NIA) is depicted graphically in Figure 10.1. Results for
BCs and Class A EPSs are presented in section 10.4. NIA results for non-Class A EPSs will be
calculated in the next (NOPR) stage of the rulemaking. Preliminary NIA results for non-Class A
EPS product classes were presented in DOEs determination for non-Class A External Power

10-2

Supplies.a However, these results were calculated using different methodologies, and thus should
not be compared with the NIA results presented here for Class A EPSs.
All calculations are contained in a Microsoft Excel workbook, which can be downloaded
from the EERE website.b The shipments analysis (see chapter 9) provides a detailed description
of the model DOE used to forecast future purchases of the considered products.

Calculation of Net Present Value


Electricity Price
National Energy
Savings ($)

Unit Energy Savings


National Energy
Savings (kWh)
National Inventory

National
Improvement Costs
Unit Improvement
Cost

Discount Rate

Net Present Value

Figure 10.1 Flow Chart Showing the Calculation of National Energy Savings and Net
Present Value
While the life-cycle cost analysis weighs the costs and savings associated with standards
for a set of representative units, the NIA considers the costs and savings associated with
standards for all products, including those in the scaled product classes. For these scaled
product classes, most inputs to the NIA could be obtained in the same manner as for the
representative product classes. However, cost data were not available for the scaled product
classes, as they were not directly analyzed in the engineering analysis. Therefore, as is done in
other rulemakings, DOE extrapolated from its knowledge of costs for the representative product
classes to the scaled product classes. In the preliminary analysis, DOE applied costs from the
EPS representative units to product classes A2, A3 and A4 as appropriate given observed
similarities between the products in these classes. Costs from the 2.5 W EPS representative unit
were applied to product classes A2 and A4, and costs from the 18 W EPS representative unit
were applied to product class A3. Costs for BC product classes 3, 6, 8, and 9 were derived from
cost data for the representative product classes as described in chapter 5.
a

See
http://www1.eere.energy.gov/buildings/appliance_standards/residential/battery_external_det_2008_non_class_a.htm
l
b
See http://www.eere.energy.gov/buildings/appliance_standards/residential/battery_external.html

10-3

10.1.1 National Inventory Accounting Model


DOE used a national inventory model to represent the number of BC and EPS units in use
during a given year that would be subject to a standard.
Unlike many other rulemakings, where an installed base of products governs shipments
(via retirement/replacement of the existing stock and installation of new stock in new homes), for
BCs and EPSs shipments govern the national inventory. DOE chose this method for calculating
national inventory based on two factors that distinguish BCs and EPSs. First, the size of the
existing inventory of products in use is not fully known. Second, DOE cannot assume that all
retired products are immediately replaced, so the number of shipments in a given year cannot be
based on the number of existing products retired in that year.
Initially there are no units in the national inventory. Each year after a standard takes
effect, new units are added to the inventory and those units that have reached the end of their
lives are removed from the inventory. DOE used two inputs to calculate the national inventory:
shipments forecasts (see chapter 9) and product class lifetime profiles, which are derived by
combining base-year shipments by application with application lifetimes (see chapter 3).
10.1.2 Calculation of Product Class Lifetime Profiles
For the preliminary analysis, DOE calculated product class lifetime profiles using the
percentage of shipments of applications within a given product class, and the lifetimes of those
applications. These values were combined to estimate the percentage of units remaining in use
for each year following the initial year in which those units were shipped.
As an example, consider a product class X with four associated applications: A, B, C,
and D. Base year shipments and lifetimes for these applications are shown in Table 10.3.
Table 10.3 Example Product Class Characteristics for Calculation of Lifetime Profile
Application

Base Year Shipments


(Units)

Percentage of Base Year


Shipments
(%)

Lifetime
(Years)

100,000

53%

25,000

13%

15,000

8%

50,000

26%

Based on these application-specific values, product class Xs lifetime profile would be as


shown in Table 10.4.
Table 10.4 Example Lifetime Profile
Percentage of Units Remaining in Use
2013
(Shipment

2014

2015

2016

10-4

2017

2018

Year)
100%

100%

92%

79%

26%

0%

In 2013, 100% of units are shipped and put into use.


At the end of 2013, after 1 year of use, none of the products have reached their lifetime,
so 0% of the units are retired, leaving 100% of the initial shipment remaining in use in
2014.
At the end of 2014, after 2 years of use, units for product C, representing 8% of the total
initial shipment, reach their lifetime and are retired, leaving 92% of the initial shipment
remaining in use in 2015.
At the end of 2015, after 3 years of use, units for product B, representing 13% of the total
initial shipment, reach their lifetime and are retired, leaving 79% of the initial shipment
remaining in use in 2016.
At the end of 2016, after 4 years of use, units for product A, representing 53% of the total
initial shipment, reach their lifetime and are retired, leaving 26% of the initial shipment
remaining in use in 2017.
At the end of 2017, after 5 years of use, units for product D, representing 26% of the total
initial shipment, reach their lifetime and are retired, leaving 0% of the initial shipment
remaining in use in 2018.
Since no units remain in use in 2018, the lifetime profile calculation is complete.
10.1.3 National Inventory Example
Table 10.5 gives an example showing how the national inventory is tracked over the first
seven years of the analysis period. This example uses as inputs:
Shipments of 100 units in 2013

Market growth of 5 percent per year

The lifetime profile shown in Table 10.4.

Table 10.5 Example National Inventory Table


Vintage

Year of Analysis
2013

2014

2015

2016

2017

2018

2019

2013

100

100

92

79

26

2014

105

105

97

83

27

2015

110

110

101

87

29

2016

116

116

107

92

2017

122

122

112

2018

128

128

2019

134

100

205

307

402

446

468

492

National
Inventory

10-5

In 2013:
100 units are shipped (vintage 2013) and are added to the national inventory.
The total national inventory is 100 units.
In 2014:
0 units from vintage 2013 are retired, leaving 100 units.

105 new units are shipped (vintage 2014), reflecting a 5% increase over the previous

year.
The total national inventory is 205 units.
In 2015:
8 units (8%) from vintage 2013 are retired, leaving 92 units.

0 units from vintage 2014 are retired, leaving 105 units.

110 units are shipped (vintage 2015).

The total national inventory is 307 units.

In 2016:
13 units (13%) from vintage 2013 are retired, leaving 79 units.

8 units (8%) from vintage 2014 are retired, leaving 97 units.

0 units from vintage 2015 are retired, leaving 110 units.

116 units are shipped (vintage 2016).

The total national inventory is 402 units.

In 2017:
53 units (53%) from vintage 2013 are retired, leaving 26 units.

14 units (13%) from vintage 2014 are retired, leaving 83 units.

9 units (8%) from vintage 2015 are retired, leaving 101 units.

0 units from vintage 2016 are retired, leaving 116 units.

122 units are shipped (vintage 2017).

The total national inventory is 446 units.

In 2018:
26 units (26%) from vintage 2013 are retired, leaving 0 units.

56 units (53%) from vintage 2014 are retired, leaving 27 units.

14 units (13%) from vintage 2015 are retired, leaving 87 units.

9 units (8%) from vintage 2016 are retired, leaving 107 units.

0 units from vintage 2017 are retired, leaving 122 units.

128 units are shipped (vintage 2018).

The total national inventory is 468 units.

In 2019:
27 units (26%) from vintage 2014 are retired, leaving 0 units.
58 units (53%) from vintage 2015 are retired, leaving 29 units.
15 units (13%) from vintage 2016 are retired, leaving 92 units.
10-6

10 units (8%) from vintage 2017 are retired, leaving 112 units.

0 units from vintage 2018 are retired, leaving 128 units.

134 units are shipped (vintage 2019).

The total national inventory is 492 units.

10.2 National Energy Savings


DOE calculated the NES associated with a standard at each CSL for BCs and EPSs. The
calculation of NES, which represents the total energy savings for a product class over the entire
analysis period, encompasses three steps:
1) The annual unit energy savings (UES) associated with a CSL is calculated as the
difference in energy consumption between an average unit in the absence of standards
(base case) and an average unit with a standard set at that CSL.
2) The UES is then multiplied by the national inventory for a given year for that product
class to obtain annual NES for that year.
3) Annual NES is then calculated for each year of the analysis period and summed.
10.2.1 Calculating National Energy Savings
For a given product class and CSL, DOE first calculated annual UES (in kWh/year) as
the difference in annual energy consumption between an average unit in the base case (UECBase)
and an average unit in the standards case (UECCSL).
Equation 10.1

UESCSL = UECBase - UECCSL

DOE then calculated the NES in a given year (NESYear-kWh) by multiplying the national
inventory in that year (NationalInventoryYear) by the UESCSL. The calculation of NESYear-kWh is
represented by the following equation:
Equation 10.2

NESYear-kWh = NationalInventoryYear UESCSL

DOE then calculated NESkWh for the entire analysis period by repeating the above
NESYear for each year, and then summing the results. The calculation is represented by the
following equation:
Equation 10.3

NESkWh = NESYear-kWh

10.2.2 Inputs to National Energy Savings Calculation


There are three inputs to the NES calculation:
o National inventory (explained in section 10.1.1)
o Unit energy savings (explained in section 10.2.2.1)
o Site-to-source conversion factor (explained in section 10.2.2.2)

10-7

10.2.2.1 Unit Energy Savings


DOE used the efficiency distributions for the base case presented in chapter 9 along with
the annual UEC values presented in chapter 7 to estimate shipment-weighted average UEC under
the base and standards cases, which were then compared against one another to give UES values
for each CSL.
As explained in chapter 9, DOE assumed that energy efficiency would not improve after
2013 in the base case. Therefore, UEC values do not vary over time, and thus UES values do not
vary over time. In addition, DOE assumed that manufacturers would respond to a standard by
improving the efficiency of underperforming products but not those that already meet or exceed
the standard, as discussed in chapter 9.
The average annual UES for each product class and CSL is shown in Table 10.6 for EPSs
and in Table 10.7 and Table 10.8 for BCs. Values are reported for all Class A EPSs, with product
class A1 broken out into four segments by nameplate output power, corresponding to the four
representative units examined in the engineering analysis. Only reference case values are shown
here.
Table 10.6. Shipment-Weighted Average Unit Energy Savings for External Power
Supplies (kWh/yr)
Product Class

Standard Level Selected


CSL 1

CSL 2

CSL 3

CSL 4

A1: 0-10.25 W

0.46

0.91

1.16

1.27

A1: 10.25-39 W

1.24

3.01

4.74

6.06

A1: 39-90 W

0.47

1.87

3.18

5.11

A1: 90-250 W

0.88

3.12

3.73

7.84

A2 (DC Output, Low Voltage)

0.30

0.74

0.89

1.02

A3 (AC Output, Basic Voltage)

1.27

3.07

4.85

6.19

A4 (AC Output, Low Voltage)

1.03

1.69

2.27

2.37

All Class A EPSs

0.64

1.67

2.49

3.40

10-8

Table 10.7. Shipment-Weighted Average Unit Energy Savings for Battery Chargers,
Manufacturer Data (kWh/yr)
Conversion
Type

Battery
Energy

<100 Wh
AC-DC
100
3000 Wh

Battery
Voltage

AC-AC

Standard Level Selected


CSL 1

CSL 2

CSL 3

Inductive
Connection

2.25

5.31

7.31

<4 V

1.57

6.31

N/A

4<10 V

1.57

5.47

N/A

10 V

5.40

8.09

N/A

<20 V

19.06

53.92

78.97

20 V

13.69

25.55

32.56

20.00

55.00

105.00

<9 V Input

0.24

0.56

N/A

9 V Input

0.18

0.49

N/A

AC Output
from
Battery

10

11.76

14.13

16.63

>3000 Wh
DC-DC

Class
ID

Table 10.8. Shipment-Weighted Average Unit Energy Savings for Battery Chargers, Test
and Teardown Data (kWh/yr)
Conversion
Type

Battery
Energy

<100 Wh
AC-DC
100
3000 Wh

Battery
Voltage

AC-AC

Standard Level Selected


CSL 1

CSL 2

CSL 3

Inductive
Connection

<4 V

4<10 V

10 V

5.41

8.11

N/A

<20 V

23.15

N/A

N/A

20 V

No Data

No Data

<9 V Input

No Data

9 V Input

No Data

AC Output
from
Battery

10

>3000 Wh
DC-DC

Class
ID

No Data
N/A

6.09*

N/A

No Data

11.76

N/A

N/A

*Savings for BC product class 2 are relative to CSL 1 because CSL 0 is not defined for this product class in the test
and teardown data.

10-9

10.2.2.2 Site-to-Source Conversion Factors


In determining national annual energy consumption, DOE initially calculated the annual
energy consumption and savings at the site in kWh. It then converted site energy savings to
primary (source) energy savings. For all product classes except BC product class 9, this
conversion was achieved by applying a site-to-source conversion factor to account for losses
associated with the generation, transmission, and distribution of electricity.
For mains electricity DOE used annual site-to-source conversion factors based on EIAs
Annual Energy Outlook 2010 (AEO 2010), Table 4. Residential Sector Key Indicators and
Consumption.1 Figure 10.2 shows the site-to-source conversion factors for each year from 2010
through 2060. Factors for each year from 2010 to 2035 were calculated by dividing the sum of
Delivered Energy [Electricity] and Electricity Related Losses by Delivered Energy [Electricity].
The site-to-source conversion factor given by AEO 2010 for 2013 is 3.169. The factor declines
gradually from 2013 to 2035 as the power system is expected to become more energy efficient
during that period. The site-to-source conversion factor is held constant at the 2035 value in later
years, which are beyond the time horizon of the AEO 2010 forecast.

Figure 10.2 Site-to-Source Conversion Factors, 2010 to 2060


Site-to-source conversion for BCs powered by automotive power sources (BC product
class 9) differs from other product classes, as the energy source for these products is gasoline,
which is then converted into electric power by the engine. Prior research uncovered has
determined that the typical efficiency in converting gasoline to electric power is 21 percent.2 In
addition, DOE previously determined that petroleum refining and distribution efficiency is 83
percent.3 Multiplying these two efficiencies together yields a net efficiency of 17.43 percent,
which is equivalent to a site-to-source conversion factor of 5.74. DOE applied this conversion
factor over the entire analysis period. DOE assumed that power for BCs in product class 9 is

10-10

provided directly by the engine, and so did not include any efficiency losses from the automotive
battery.
10.3 Net Present Value
DOE calculated the NPV of the increased product costs and reduced operating costs
associated with a standard at each CSL for BCs and EPSs. The calculation of NPV, which
represents the present value of the difference between consumer savings and costs over the entire
analysis period, encompasses six steps:
1. The NESYear-kWh site electricity savings values are converted to energy cost savings by
multiplying by forecast electricity rates, with each year discounted to present value.
2. The unit improvement costs (UIC) associated with a CSL is calculated as the difference
in cost between an average unit in the base case and an average unit in the standards case.
3. The UIC is then multiplied by the shipments for a given shipment year for that product
class to obtain an annual National Improvement Cost (NICYear).
4. The calculation of the annual NICYear is then repeated for each year of the shipment
forecast period, with each year discounted to present value.
5. Present value NICYear is then subtracted from present value NESYear-$ for each year of the
analysis period, yielding net present value for each year (NPVYear).
6. NPVYear values are then summed to yield total NPV.
10.3.1 Calculating Net Present Value
DOE first converted site national energy savings for a given year (NESYear-kWh) to present
value energy cost savings (in 2009$) by first multiplying by the projected fuel price for that year
(FuelPriceYear) and then multiplying the product by a discount factor for that year (DFYear), as
described by the following equation. This process was repeated for each year.
Equation 10.4

NESYear-$ = NESYear-kWh FuelPriceYear DFYear

Next for a given product class and CSL, DOE first calculated UIC as the difference in
unit cost between an average unit in the base case (AvgUnitCostBase) and an average unit in the
standards case (AvgUnitCostCSL).
Equation 10.5

UICCSL = AvgUnitCostBase AvgUnitCostCSL

DOE then calculated the present value of NIC (in 2009$) for a given year (NICYear) by
multiplying the shipments for a given year for the product class (ShipmentsYear) by the UICCSL of
that product class, and multiplying the product by a discount factor for that year (DFYear). The
calculation of NICYear is represented by the following equation. This process was repeated for
each year.
Equation 10.6

NICYear = ShipmentsYear UICCSL DFYear

DOE calculated NPV for each year as the difference between NES$-Year and NICYear, with
positive values indicating cost effectiveness:

10-11

Equation 10.7

NPVYear = NES$-Year - NICYear

Finally, DOE summed NPVYear values over all years to obtain NPV.
Equation 10.8

NPV = NPVYear

10.3.2 Inputs to Net Present Value Calculation


There are six inputs to the NPV calculation:
Shipments (explained in chapter 9)

National energy savings (explained in section 10.2)

Electricity prices (explained in section 10.3.2.1)

Gasoline prices (explained in section 10.3.2.2)

Unit improvement costs (explained in section 10.3.2.3)

Discount rates (explained in section 10.3.2.4)

10.3.2.1 Electricity Prices


DOE used the methodology described in chapter 8 for forecasting electricity prices for
residential and commercial consumers. In the preliminary analysis, DOE assumed all energy
consumption and savings would take place in the residential sector and used a trend function to
extend the electricity price forecast used in the LCC analysis from 2030 until the end of the
analysis period.
10.3.2.2 Gasoline Prices
For BCs powered by automotive power sources (BC product class 9), DOE calculated
consumer energy cost savings using gasoline prices rather than electricity prices. To assign a
price to the energy consumed by these BCs, DOE first converted the electricity consumed by the
BC (in kWh) into the equivalent quantity of gasoline (in gallons) by using an energy equivalence
conversion factor of 33.705 kWh per gallon of gasoline.3 DOE then divided the resulting gallons
by average automobile energy conversion efficiency (21 percent, from section 10.2.2.2). To
convert these gasoline savings into consumer cost savings, DOE multiplied the resulting gallons
of gasoline by the gasoline price.
DOE used the gasoline price forecast in AEO 2010, and converted prices from 2008$ to
2009$. As with the electricity price forecasts, DOE used an Excel trend function to extend the
gasoline price forecast from 2030 to 2060.
10.3.2.3 Unit Improvement Costs
DOE used the efficiency distributions for the base case presented in chapter 9, the
manufacturer selling prices presented in chapter 5, and markups and sales tax from chapter 6 to
calculate the UIC in each standards case. Manufacturers are assumed to respond to a standard by
improving the efficiency of underperforming products but not those that already meet or exceed
the standard, as discussed in chapter 9. Average unit improvement costs in each scenario are
shown in Table 10.9 for EPSs and Table 10.10 and Table 10.11 for BCs.

10-12

In the preliminary analysis, DOE applied costs from the EPS representative units to
product classes A2, A3 and A4 as appropriate given observed similarities between the products
in these classes. Costs from the 2.5 W EPS representative unit were applied to product classes
A2 and A4, and costs from the 18 W EPS representative unit were applied to product class A3.
Similarly, DOE derived cost estimates for scaled BC product classes 3, 6, 8, and 9 from its cost
estimates for the representative product classes. These calculations are discussed in chapter 5.
Table 10.9 Shipment-Weighted Average Unit Improvement Costs for External Power
Supplies (2009$) Manufacturer Data
Standard Level Selected
Product Class

CSL 1

CSL 2

CSL 3

CSL 4

A1: 0-10.25 W

$0.02

$0.26

$0.38

$0.54

A1: 10.25-39 W

$0.15

$0.28

$0.77

$1.40

A1: 39-90 W

$0.12

$0.69

$1.91

$3.81

A1: 90-250 W

$0.31

$0.99

$1.23

$6.31

A2 (DC Output, Low Voltage)

$0.02

$0.27

$0.40

$0.56

A3 (AC Output, Basic Voltage)

$0.15

$0.28

$0.78

$1.40

A4 (AC Output, Low Voltage)


All Class A EPSs

$0.02

$0.25

$0.36

$0.51

$0.08

$0.37

$0.80

$1.57

Table 10.10 Shipment-Weighted Average Unit Improvement Costs for Battery Chargers
(2009$) Manufacturer Data
Conversion
Type

Battery
Energy

Standard Level Selected


Class ID

CSL 1

CSL 2

CSL 3

Inductive
Connection

$0.16

$0.43

<4 V

$0.28

$15.81

$0.64
N/A

4<10 V

$0.87

$10.98

N/A

<100 Wh

10 V

$1.51

$4.27

N/A

1003000
Wh

<20 V

$5.72

$37.89

$130.21

20 V

$12.69

$54.57

$146.76

$11.67

$44.62

<9 V Input

$0.46

$1.54

$140.28
N/A

DC-DC

9 V Input

$0.35

$3.25

N/A

AC-AC

AC Output
from Battery

10

$1.50

$1.96

$2.53

AC-DC

Battery Voltage

>3000 Wh

10-13

Table 10.11 Shipment-Weighted Average Unit Improvement Costs for Battery Chargers
(2009$) Test and Teardown Data
Conversion
Type

Battery
Energy

<100 Wh
AC-DC
1003000
Wh

Battery Voltage

AC-AC

Standard Level Selected


CSL 1

CSL 2

CSL 3

Inductive
Connection

<4 V

4<10 V

10 V

$0.78

$1.38

N/A

<20 V

$2.06

N/A

N/A

20 V

No Data

No Data

<9 V Input

No Data

9 V Input

No Data

AC Output
from Battery

10

>3000 Wh
DC-DC

Class ID

No Data
N/A

$2.93*

N/A

No Data

-$0.65

N/A

N/A

*Costs for BC product class 2 are relative to CSL 1 because CSL 0 is not defined for this product class in the test
and teardown data.

Note: In the NIA workbook, manufacturer selling prices (MSPs) are presented for both
EPSs and BCs. While prices for EPSs are marginal (adjusted so that the price of a unit at CSL 0
is zero), and prices for BCs are absolute, the NIA calculation of UIC employs the same
methodology for both products.
10.3.2.4 Discount Rates
To calculate NPV, DOE discounted future consumer costs and savings to the present day
(2010) using discount rates of 3 percent and 7 percent. These discount rates are specified by the
U.S. Office of Management and Budget (OMB) (OMB, Circular A-4: Regulatory Analysis,
2003).
10.4 Results
DOE estimated NES and NPV for BC standards separately from EPS standards. Results
for the two products should not be added to one another because many BCs incorporate an EPS.
One way to improve the efficiency of these BCs is to improve the efficiency of the EPS part. The
resulting savings are counted in both the BC analysis and in the separate EPS analysis. Thus, due
to this overlap, combining BC and EPS NES (or NPV) estimates would overstate savings
resulting from improving BCs and EPSs.

10-14

10.4.1 National Energy Savings


DOEs preliminary analysis finds that new standards for BCs and amended standards for
EPSs could produce cumulative energy savings over the period of analysis of as much as
3.5 quadrillion Btu (quads) and 1.5 quads, respectively, in the case where energy savings are
maximized. In the case where consumer benefits are maximized, new standards for BCs and
amended standards for EPSs could produce cumulative energy savings over the period of
analysis of 1.4 quadrillion Btu (quads) and 1.0 quads, respectively.
Results for each standards case are shown in Table 10.12 for EPSs, and Table 10.13 and
Table 10.14 for BCs. In each cell, results from the reference savings case are shown in the top of
the cell, with results from the low and high savings cases shown in parentheses in the bottom of
the cell. Low and high results are only shown for product classes for which energy savings vary
between scenarios.
Table 10.12 National Energy Savings Potential from Amended Standards for External
Power Supplies (Trillion Btu)
Product Class
A1: 0-10.25 W
A1: 10.25-39 W
A1: 39-90 W
A1: 90-250 W
A2 (DC Output, Low
Voltage)
A3 (AC Output, Basic
Voltage)
A4 (AC Output, Low
Voltage)
All Class A EPSs

Standard Level Selected


CSL 1

CSL 2

CSL 3

CSL 4

57
(46 to 58)
132
(119 to 145)
34
(31 to 43)
10
(9 to 15)

114
(99 to 142)
319
(287 to 352)
137
(128 to 167)
35
(32 to 54)

144
(127 to 180)
504
(453 to 555)
232
(217 to 286)
42
(38 to 64)

158
(140 to 202)
643
(578 to 710)
374
(345 to 466)
88
(78 to 137)

32

81

96

111

26

64

101

128

4
296
(267 to 325)

7
756
(697 to 866)

9
1128
(1041 to 1291)

10
1512
(1389 to 1763)

10-15

Table 10.13 National Energy Savings Potential from New Standards for Battery
Chargers, Manufacturer Data (Trillion Btu)
Conversion
Type

Battery
Energy

<100 Wh
AC-DC
100
3000 Wh

Battery
Voltage

CSL 1

CSL 2

CSL 3

Inductive
Connection

22

52

71

<4 V

4<10 V

10 V

<20 V

20 V

1596
(1276 to 1792)
206
813
(694 to 1149)
250
(109 to 250)
76
26
(12 to 43)
50

366
(136 to 366)
97
49
(24 to 81)
N/A

<9 V Input

397
(338 to 422)
59
543
(465 to 767)
88
(19 to 88)
41
9
(5 to 16)
21

9 V Input

10

26

N/A

AC Output
from
Battery

10

177

213

250

>3000 Wh
DC-DC

AC-AC

Standard Level Selected

Class
ID

10-16

N/A
N/A
N/A

Table 10.14 National Energy Savings Potential from New Standards for Battery
Chargers, Test and Teardown Data (Trillion Btu)
Conversion
Type

Battery
Energy

<100 Wh
AC-DC
100
3000 Wh

Battery
Voltage

AC-AC

CSL 1

CSL 2

Inductive
Connection

<4 V

4<10 V

10 V

<20 V

20 V

No Data

No Data

<9 V Input

No Data

9 V Input

No Data

AC Output
from
Battery

10

>3000 Wh
DC-DC

Standard Level Selected

Class
ID

CSL 3

No Data
N/A
544
(478 to 773)
107
(63 to 107)

177

1541*
(1273 to 1733)
No Data
815
(722 to 1164)
N/A

N/A

N/A

N/A
N/A

N/A

*Savings for BC product class 2 are relative to CSL 1 because CSL 0 is not defined for this product class in the test
and teardown data.

10.4.2 Annual Costs and Savings


To illustrate the basic inputs to the NPV calculations, Figure 10.3 presents the
undiscounted NIC and NES at CSL 1 for AC-DC basic voltage Class A EPSs with nameplate
output power up to 10.25 watts. The figure also shows the net consumer savings, which is the
difference between the NIC and NES for each year. DOE can create figures like this one for each
standards case.

10-17

Figure 10.3 Reference Case Annual Undiscounted Consumer Costs and Savings from a
Standard at CSL 1 for AC-DC Basic Voltage Class A EPSs with Nameplate Output
Power up to 10.25 Watts
10.4.3 Net Present Value
The tables in this section present DOEs preliminary estimates of the national NPV of
total consumer costs and savings expected to result from standards at specific efficiency levels
for BCs and EPSs.
DOEs preliminary analysis finds that new standards for BCs and amended standards for
EPSs could yield net consumer savings over the period of analysis of $4.5 billion and $3.4
billion, respectively, in the case where consumer benefits are maximized. In the case where
energy savings are maximized, new standards for BCs and amended standards for EPSs could
yield net consumer savings over the period of analysis of as much as -$70 billion and $0.25
billion, respectively. These results assume the reference case electricity price forecast.
Results for EPSs are shown first, using discount rates of 3 percent (Table 10.15) and 7
percent (Table 10.16). The corresponding values for BCs are shown in Table 10.17 and Table
10.18 (derived from manufacturer data) and Table 10.19 and Table 10.20 (derived from test and
teardown data). In each cell, results from the reference savings case are shown in the top of the
cell, with results from the low and high savings cases shown in parentheses in the bottom of the
cell. Low and high results are only shown for product classes for which NPV varies between
scenarios.

10-18

Table 10.15 Net Present Value of Consumer Savings from Amended Standards for
External Power Supplies at a 3 Percent Discount Rate, Manufacturer Data (2009$
millions)
Product Class

Standard Level Selected


CSL 1

CSL 2

CSL 3

CSL 4

332
(261 to 335)
637
(391 to 807)

315
(212 to 496)
1663
(1164 to 2027)

322
(202 to 562)
2155
(1007 to 2921)

A1: 39-90 W

90
(24 to 172)

117
(-199 to 442)

-606
(-1426 to 96)

A1: 90-250 W

-2
(-27 to 47)

18
(-56 to 178)

10
(-81 to 205)

160
(29 to 457)
2186
(281 to 3402)
-1785
(-3415 to 476)
-768
(-1157 to 183)

174

106

22

-136

138

348

492

560

26

32

42

1394
(986 to 1698)

2600
(1608 to 3631)

2437
(258 to 4340)

38
254
(-3800 to
3662)

A1: 0-10.25 W
A1: 10.25-39 W

A2 (DC Output, Low


Voltage)
A3 (AC Output, Basic
Voltage)
A4 (AC Output, Low
Voltage)
All Class A EPSs

Note: Reference case electricity prices are assumed.

Table 10.16 Net Present Value of Consumer Savings from Amended Standards for
External Power Supplies at a 7 Percent Discount Rate, Manufacturer Data (2009$
millions)
Product Class
A1: 0-10.25 W
A1: 10.25-39 W
A1: 39-90 W
A1: 90-250 W
A2 (DC Output, Low
Voltage)
A3 (AC Output, Basic
Voltage)
A4 (AC Output, Low
Voltage)

Standard Level Selected


CSL 1

CSL 2

CSL 3

CSL 4

165
(130 to 164)
311
(180 to 400)

138
(88 to 224)
819
(556 to 1008)

133
(75 to 247)
1035
(422 to 1439)

41
(6 to 82)
-3
(-16 to 22)

29
(-142 to 195)
3
(-37 to 84)

-392
(-836 to -29)
-3
(-52 to 96)

40
(-24 to 183)
1012
(-10 to 1658)
-1074
(-1957 to 393)
-438
(-649 to -136)

86

37

-14

-104

63

161

223

248

13

16

20

18

10-19

All Class A EPSs

675
(462 to 830)

1202
(679 to 1724)

1004
(-161 to 1982)

-298
(-2478 to
1473)

Note: Reference case electricity prices are assumed.

Table 10.17 Net Present Value of Consumer Savings from New Standards for Battery
Chargers at a 3 Percent Discount Rate, Manufacturer Data (2009$ millions)
Conversion
Type

Battery
Energy

<100 Wh

AC-DC
100
3000 Wh

Battery
Voltage

AC-AC

CSL 1

CSL 2

CSL 3
382
(369 to 391)

-39
1291
(265 to 2988)

281
(272 to 287)
-57441
(-74744 to 48000)
-3957
-980
(-3283 to 2023)

274
(-213 to 306)

-277
(-1503 to -64)
-603
-37
(-135 to 72)
-1970

-4026
(-6658 to 3293)
-2277
-315
(-525 to -90)
N/A

Inductive
Connection

121
(118 to 123)

<4 V

1360
(718 to 1654)

4<10 V

10 V

<20 V

20 V

N/A
N/A
N/A

<9 V Input

0.05
7
(-27 to 45)
-552

9 V Input

29

-658

N/A

10

898
(857 to 923)

1056
(1003 to 1089)

1211
(1142 to
1253)

>3000 Wh
DC-DC

Standard Level Selected

Class
ID

AC Output
from
Battery

Note: Reference case electricity prices are assumed.

10-20

Table 10.18 Net Present Value of Consumer Savings from New Standards for Battery
Chargers at a 7 Percent Discount Rate, Manufacturer Data (2009$ millions)
Conversion
Type

Battery
Energy

<100 Wh

AC-DC
100
3000 Wh

Battery
Voltage

AC-AC

CSL 1

CSL 2

CSL 3
187
(179 to 191)

N/A

Inductive
Connection

59
(57 to 60)

<4 V

641
(310 to 783)

137
(132 to 140)
-31855
(-41224 to 26790)

4<10 V

10 V

-40
555
(31 to 1361)

-2229
-762
(-1967 to 696)

<20 V

115
(-124 to 132)

-251
(-866 to -134)

N/A
-2349
(-3699 to 1948)

20 V

-21
0.2
(-16 to 16)

-370
-30
(-76 to 15)

-1298
-191
(-292 to -96)

>3000 Wh
DC-DC

Standard Level Selected

Class
ID

N/A

<9 V Input

-308

-1090

N/A

9 V Input

-384

N/A

AC Output
from
Battery

10

425
(402 to 438)

498
(469 to 516)

569
(531 to 592)

Note: Reference case electricity prices are assumed.

10-21

Table 10.19 Net Present Value of Consumer Savings from New Standards for Battery
Chargers at a 3 Percent Discount Rate, Test and Teardown Data (2009$ millions)
Conversion
Type

Battery
Energy

<100 Wh
AC-DC
100
3000 Wh

Battery
Voltage

AC-AC

CSL 1

CSL 2

Inductive
Connection

<4 V

4<10 V

10 V

<20 V

20 V

No Data

No Data

<9 V Input

No Data

9 V Input

No Data

AC Output
from
Battery

10

>3000 Wh
DC-DC

Standard Level Selected

Class
ID

CSL 3

No Data
N/A
2351
(1949 to 3744)
570
(293 to 570)

1232

-2638*
(-4342 to -1467)
No Data
3225
(2661 to 5345)
N/A

N/A

N/A

N/A
N/A

N/A

*Values for BC product class 2 are relative to CSL 1 because CSL 0 is not defined for this product class in the test

and teardown data.

Note: Reference case electricity prices are assumed.

10-22

Table 10.20 Net Present Value of Consumer Savings from New Standards for Battery
Chargers at a 7 Percent Discount Rate, Test and Teardown Data (2009$ millions)
Conversion
Type

Battery
Energy

<100 Wh
AC-DC
100
3000 Wh

Battery
Voltage

AC-AC

CSL 1

CSL 2

Inductive
Connection

<4 V

4<10 V

10 V

<20 V

20 V

No Data

No Data

<9 V Input

No Data

9 V Input

No Data

AC Output
from
Battery

10

>3000 Wh
DC-DC

Standard Level Selected

Class
ID

CSL 3

No Data
N/A
1136
(950 to 1773)
269
(135 to 269)

608

-1845
(-2689 to -1304)
No Data
1539
(1280 to 2509)
N/A

N/A

N/A

N/A
N/A

N/A

*Values for BC product class 2 are relative to CSL 1 because CSL 0 is not defined for this product class in the test

and teardown data.

Note: Reference case electricity prices are assumed.

10-23

REFERENCES
1

Energy Information Administration. Annual Energy Outlook 2010.

http://www.eia.doe.gov/oiaf/aeo/aeoref_tab.html

2
Bradfield, M. Improving Alternator Efficiency Measurably Reduces Fuel Costs. 2008. Remy International, Inc:

Pendleton, IN.

3
Title 10, Code of Federal Regulations, Part 474, Subpart C Electric and Hybrid Vehicle Research, Development,
and Demonstration Program. June 12, 2000. (http://www.gpo.gov/fdsys/pkg/FR-2000-06-12/html/00-14446.htm)

10-24

CHAPTER 11. LIFE-CYCLE COST SUBGROUP ANALYSIS

TABLE OF CONTENTS

11.1
11.2

INTRODUCTION ......................................................................................................... 11-1

PURCHASE PRICE IMPACTS .................................................................................... 11-1

11-i

CHAPTER 11. LIFE-CYCLE COST SUBGROUP ANALYSIS

11.1

INTRODUCTION

The life-cycle cost (LCC) subgroup analysis evaluates impacts on any identifiable groups
or customers who may be disproportionately affected by any national energy efficiency standard
level. The U.S. Department of Energy (DOE) will conduct this analysis as one of the analyses for
the notice of proposed rulemaking. DOE will accomplish this, in part, by analyzing the LCC and
payback periods (PBPs) for those customers that fall into any identifiable groups. DOE plans to
evaluate variations in regional energy prices and variations in energy use that might affect the net
present value of a standard to customer subpopulations. To the extent possible, DOE will obtain
estimates of each input parameters variability and will consider this variability in its calculation
of customer impacts. DOE plans to perform sensitivity analyses to consider how differences in
energy use will affect subgroups of customers.
DOE will determine the impact on customer subgroups using the LCC spreadsheet
model, which allows for different data inputs. The standard life-cycle cost and payback period
analysis (described in the preliminary Technical Support Document (TSD) chapter 8) focuses on
residential and commercial consumers that use battery chargers (BC) and external power
supplies (EPS). DOE can use the LCC spreadsheet model to analyze the LCC for any subgroup
by sampling only that subgroup. Model inputs used by the Department to determine LCC and
PBPs are described in detail in the life-cycle cost and payback period analysis (see preliminary
TSD chapter 8).
In the case of BCs and EPSs, some possible subgroups DOE may choose to consider are:
(1) small businesses (i.e., those with low annual revenues), (2) consumers with a high marginal
electricity rate, (3) low-income consumers, and (4) consumers of specific applications within a
representative unit. These subgroups may experience different economic conditions than the
average owner of a BC or EPS. Small businesses are likely to have higher borrowing costs and
cost of capital than larger commercial firms. Consumers with a high marginal electricity rate
may experience very different operating costs based on new standards when compared to the
national average electricity rate.
11.2

PURCHASE PRICE IMPACTS

DOE will be especially sensitive to increases in the purchase price of the equipment due
to new standards, to avoid negative impacts on identifiable population groups that may not be
able to afford significant increases in equipment price. For such customers that are sensitive to
price increases, increases in first costs of a product can preclude the purchase of a new model of
that product. As a result, some customers may retain products past their useful life. These older
products are generally less efficient to begin with, and their efficiency may deteriorate further if
they are retained beyond their useful life. Increases in first cost also can preclude the purchase
and use of a product altogether, resulting in a potentially large loss of utility to the customer.

11-1

CHAPTER 12 PRELIMINARY MANUFACTURER IMPACT ANALYSIS

TABLE OF CONTENTS

12.1
12.2

12.3

INTRODUCTION........................................................................................................ 12-1

METHODOLOGY ...................................................................................................... 12-1

12.2.1 Phase I: Industry Profile.................................................................................... 12-2

12.2.2 Phase II: Industry Cash-Flow Analysis and Interview Guide........................... 12-2

12.2.2.1 Industry Cash-Flow Analysis........................................................... 12-2

12.2.2.2 Interview Guide ............................................................................... 12-2

12.2.3 Phase III: Subgroup Analysis............................................................................ 12-2

12.2.3.1 Manufacturer Interviews .................................................................. 12-3

12.2.3.2 Revised Industry Cash-Flow Analysis............................................. 12-3

12.2.3.3 Manufacturer Subgroup Analysis .................................................... 12-3

12.2.3.4 Competitive Impact Assessment...................................................... 12-3

12.2.3.5 Manufacturing Capacity Impact ...................................................... 12-4

12.2.3.6 Employment Impact......................................................................... 12-4

12.2.3.7 Cumulative Regulatory Burden ....................................................... 12-4

PRELIMINARY MANUFACTURER IMPACT ANALYSIS OVERVIEW......... 12-4

12.3.1 Industry Overview ............................................................................................. 12-4

12.3.1.1 Industry Cost Structure .................................................................... 12-5

12.3.1.2 Inventory Levels .............................................................................. 12-6

12.3.2 Interview Topics and Preliminary Findings - EPS ............................................ 12-6

12.3.2.1 Market Shares and Industry Consolidation...................................... 12-7

12.3.2.2 Production and Product Mix ............................................................ 12-7

12.3.2.3 Conversion Costs ............................................................................. 12-8

12.3.2.4 Cumulative Regulatory Burden ....................................................... 12-8

12.3.2.5 Other Key Issues .............................................................................. 12-8

12.3.3 Interview Topics and Preliminary Findings - BC .............................................. 12-9

12.3.3.1 Market Shares and Industry Consolidation...................................... 12-9

12.3.3.2 Production and Product Mix ............................................................ 12-9

12.3.3.3 Conversion Costs ........................................................................... 12-10

12.3.3.4 Cumulative Regulatory Burden ..................................................... 12-10

12.3.3.5 Other Key Issues ............................................................................ 12-10

12-i

LIST OF TABLES
TABLE 12.1 ALL OTHER MISCELLANEOUS ELECTRICAL EQUIPMENT AND

COMPONENT MANUFACTURING INDUSTRY EMPLOYMENT AND EARNINGS...... 12-5

TABLE 12.2 ALL OTHER MISCELLANEOUS ELECTRICAL EQUIPMENT AND

COMPONENT MANUFACTURING INDUSTRY MATERIAL AND PAYROLL COSTS . 12-5

TABLE 12.3 ALL OTHER MISCELLANEOUS ELECTRICAL EQUIPMENT AND

COMPONENT MANUFACTURING INDUSTRY END-OF-YEAR INVENTORY ............. 12-6

TABLE 12.4 ALL OTHER MISCELLANEOUS ELECTRICAL EQUIPMENT AND

COMPONENT MANUFACTURING INDUSTRY FULL PRODUCTION CAPACITY

UTILIZATION RATES............................................................................................................. 12-6

12-ii

CHAPTER 12 PRELIMINARY MANUFACTURER IMPACT ANALYSIS

12.1

INTRODUCTION

The purpose of the manufacturer impact analysis (MIA) is to identify and quantify the
likely impacts of amended energy conservation standards on manufacturers. In the Notice of the
Proposed Rulemaking (NOPR), DOE considers a wide range of quantitative and qualitative
industry impacts that might occur due to an amended energy conservation standard. For example,
a particular standard level could require changes in manufacturing practices, equipment, raw
materials, etc. DOE fully analyzes these impacts during the NOPR stage of analysis.
DOE announced changes to the preliminary analysis MIA format through a report issued
to Congress on January 31, 2006 (as required by section 141 of the Energy Policy Act of 2005
(EPACT 2005), entitled Energy Conservation Standards Activities.1 As a result, DOE collects,
evaluates, and reports preliminary MIA information in the preliminary analysis (as opposed to
waiting for the NOPR stage). Such preliminary information includes market data, market shares,
industry consolidation, product mix, key issues, conversion costs, foreign competition, and
cumulative regulatory burden information, if available. DOE solicits this information during the
preliminary manufacturer interviews and reports the results in this chapter. Appendix 12A
includes a copy of the interview guide that DOE distributed to manufacturers.
To the extent appropriate for this rulemaking, DOE plans to apply the methodology
described below to evaluate amended energy conservation standards for battery chargers (BCs)
and external power supplies (EPSs).
12.2

METHODOLOGY

DOE conducts the MIA in three phases. In Phase I, DOE creates an industry profile to
characterize the industries and conducts a preliminary MIA to identify important issues that
require consideration. Section 12.2.1 of this chapter presents initial findings of the Phase I
analysis. In Phase II, DOE prepares an industry cash flow model and a detailed interview
questionnaire to guide subsequent discussions with manufacturers. In Phase III, DOE interviews
manufacturers and assesses the impacts of amended energy conservation standards both
quantitatively and qualitatively. DOE assesses industry and subgroup cash flow impacts and
industry net present value (INPV) using the Government Regulatory Impact Model (GRIM).
DOE also assesses impacts on competition, manufacturing capacity, employment, and regulatory
burden based on manufacturer interviews and discussions. The Federal Register NOPR and
technical support document (TSD) present results of the Phase II and III analyses.

This report is available on the DOE website at:


www.eere.energy.gov/buildings/appliance_standards/2006_schedule_setting.html.

12-1

12.2.1 Phase I: Industry Profile


In Phase I of the MIA, DOE collects pertinent qualitative and quantitative financial and
market information. This includes data on wages, employment, industry costs, and capacity
utilization rates for manufacturers of BCs and EPSs. Sources of information include reports
published by industry groups, trade journals, the U.S. Census Bureau, and Securities Exchange
Commission (SEC) 10-K filings. In addition, DOE relies on information from its market and
technology assessment, engineering analysis, life-cycle cost analysis, shipments analysis, and
product price determination to characterize the BC and EPS manufacturing industries.
12.2.2 Phase II: Industry Cash-Flow Analysis and Interview Guide
In Phase II, DOE performs a preliminary industry cash-flow analysis and prepares written
guidelines for interviewing manufacturers.
12.2.2.1

Industry Cash-Flow Analysis

DOE uses the GRIM to analyze the financial impacts of amended energy conservation
standards. Amended energy conservation standards may require additional investment, higher
production costs, and could affect revenue through higher prices and, potentially, lower
shipments. The GRIM uses several financial parameters to determine a series of annual cash
flows for the year that amended energy conservation standards become effective and for several
additional years. These factors include annual expected revenues, costs of sales, selling and
general administration expenses, taxes, and capital expenditures. Inputs to the GRIM include
those financial parameters, manufacturing costs, shipment forecasts, and price forecasts
developed in other analyses. The financial information is developed from publicly available data
and confidentially submitted manufacturer information. DOE compares the GRIM results for the
standards case at each trial standard level against the results for the base case in which no
amended energy conservation standards are in place. The financial impact of amended energy
conservation standards is the difference between the two sets of discounted annual cash flows.
12.2.2.2

Interview Guide

DOE conducts interviews with manufacturers to gather information on the effects of


amended energy conservation standards on revenues, costs, direct employment, capital assets,
and industry competitiveness. Before the interviews, which occur in Phase III, DOE distributes
an interview guide to help identify the impacts of amended energy conservation standards on
individual manufacturers or subgroups of manufacturers. Interview guide topics include:
production costs; shipment projections; market share; product mix; conversion costs; markups
and profitability; competition; manufacturing capacity; cumulative regulatory burden; and other
relevant topics.
12.2.3 Phase III: Subgroup Analysis
Phase III activities take place after publication of the preliminary analysis. These
activities include manufacturer interviews; revision of the industry cash-flow analysis; a
manufacturer subgroup analysis; and an assessment of the impacts on industry competition,
manufacturing capacity, direct employment, and cumulative regulatory burden.
12-2

12.2.3.1

Manufacturer Interviews

DOE conducts detailed interviews with manufacturers to gain insight into the potential
impacts of amended energy conservation standards on sales, direct employment, capital assets,
and industry competitiveness. The interview process is critical to the MIA because it provides an
opportunity for interested parties to privately express their views on important issues. Interviews
are scheduled well in advance to provide every opportunity for stakeholders to be available for
comment. Although a written response to the questionnaire is acceptable, DOE prefers
interactive interviews, which help clarify responses and provide the opportunity to identify
additional issues not specifically addressed in the interview questionnaire. A non-disclosure
agreement allows DOE to consider confidential or sensitive information in its decision-making
process. Confidential information will not be made available in the public record. At most,
sensitive or confidential information may be aggregated and presented in industry-wide
representations.
DOE uses information gathered during manufacturer interviews to supplement the
information gathered in Phase I and the cash flow analysis performed in Phase II.
12.2.3.2

Revised Industry Cash-Flow Analysis

As discussed, DOE requests information about profitability impacts, changes in capital


expenditures, and other manufacturing impacts during the interview process. DOE revises its
industry cash flow model based on the feedback it receives in written comments and during
interviews.
12.2.3.3

Manufacturer Subgroup Analysis

Using average cost assumptions to develop an industry cash flow estimate will not
adequately assess differential impacts among manufacturer subgroups. Smaller manufacturers,
niche players, and manufacturers exhibiting a cost structure that differs greatly from the industry
average could be more negatively affected. Ideally, DOE would consider the impact on every
firm individually; however, it typically uses the results of the industry characterization to group
manufacturers with similar characteristics. During the interviews, DOE discusses the potential
subgroups that have been identified for the analysis. DOE asks manufacturers and other
interested parties to suggest what subgroups or characteristics are most appropriate for the
analysis.
12.2.3.4

Competitive Impact Assessment

Section 342 (6)(B)(i)(V) of the Energy Policy Act of 1992 (EPCA) directs DOE to
consider any lessening of competition likely to result from the imposition of standards. EPCA
further directs the U.S. Attorney General to determine the impacts, if any, of any decrease in
competition. DOE makes a determined effort to gather and report firm-specific financial
information and impacts. DOE bases the competitive impact assessment on manufacturer cost
data and other information collected from interviews. When assessing competitive impacts,
DOEs interviews generally focus on assessing asymmetrical cost increases, the potential
increase in business risks from an increased proportion of fixed costs, and potential barriers to
12-3

market entry (e.g., proprietary technologies). The competitive analysis may also focus on
assessing any differential impacts on smaller manufacturers.
12.2.3.5

Manufacturing Capacity Impact

One of the significant outcomes of amended energy conservation standards can be the
obsolescence of existing manufacturing assets, including tooling and other investments. The
manufacturer interview guide presents a series of questions to help identify impacts on
manufacturing capacity, specifically capacity utilization and plant location decisions in North
America with and without amended energy conservation standards. The interview guide also
addresses the ability of manufacturers to upgrade or remodel existing facilities to accommodate
the new requirements; the nature and value of stranded assets, if any; and estimates for any onetime restructuring or other charges, where applicable.
12.2.3.6

Employment Impact

The impact of amended energy conservation standards on employment is an important


consideration in the rulemaking process. To assess how domestic employment patterns might be
affected, the interview process explores current employment trends in the BC and EPS industries
and solicits manufacturer views on changes in employment patterns that may result from new or
amended standards. The employment impacts section of the interview guide focuses on current
employment levels at production facilities, expected future employment levels with and without
an amended energy conservation standard, differences in workforce skills, and employee
retraining.
12.2.3.7

Cumulative Regulatory Burden

DOE seeks to mitigate the overlapping effects on manufacturers of energy conservation


standards and other regulatory actions. DOE analyzes and considers the impact on manufacturers
of multiple, product-specific regulatory actions.
12.3

PRELIMINARY MANUFACTURER IMPACT ANALYSIS OVERVIEW

During the preliminary analysis phase, DOE conducted a preliminary evaluation of the
impact of potential new and amended energy conservation standards on the BC and EPS
industries.
The primary sources of information for this analysis are the U.S. Census, industry
reports, and interviews with manufacturers of EPSs and BCs, conducted in late 2009 and early
2010. To maintain confidentiality, DOE only reports aggregated information here. DOE does not
disclose company-specific information, nor does it identify the individual manufacturers that
disclosed information.
12.3.1 Industry Overview
The following section summarizes publicly available industry data.

12-4

12.3.1.1

Industry Cost Structure

DOE is unaware of any publicly available industry-wide cost data specific to only
manufacturers of BCs and EPSs. BC and EPS manufacturing is classified as a subset under the
North American Industry Classification System (NAICS) code 335999 (All Other Miscellaneous
Electrical Equipment and Component Manufacturing). Therefore, DOE presents the data below
as a broader industry proxy for the BC and EPS industries, which, in combination with
information gained in interviews, inform DOEs analysis of the industry cost structures. For
simplicity, DOE will refer to these broader categories by the products they represent, namely
BCs and EPSs. DOE obtained the below data from U.S. Census Bureau, Annual Survey of
Manufacturers, Statistics for Industry Groups and Industries from 2002-2006.
Table 12.1 presents the employment levels and earnings from 2002 to 2006. The statistics
illustrate approximately a 40-percent decrease in production workers from 2002 to 2006. The
statistics show a 36-percent decrease in the overall number of employees from 2002 to 2006.
Table 12.1 All Other Miscellaneous Electrical Equipment and Component Manufacturing
Industry Employment and Earnings
Year
2002
2003
2004
2005
2006

Production Workers

All Employees

Annual Payroll
($000s)

30,475
49,388
1,969,580
26,571
43,078
1,862,507
25,394
42,123
1,962,922
18,429
32,376
1,558,132
18,385
31,786
1,589,923
U.S. Census Bureau. Annual Survey of Manufacturers, 2002-2006

Table 12.2 presents the costs of materials and industry payroll as a percentage of
shipment value from 2002 to 2006. The cost of materials as a percentage of shipment value has
steadily decreased from 2002 to 2006. During the 2002 to 2006 period, the cost of payroll for
production workers and the cost of total payroll have declined by 33.0 percent and 23.1 percent,
respectively.
Table 12.2 All Other Miscellaneous Electrical Equipment and Component Manufacturing
Industry Material and Payroll Costs
Year
2002
2003
2004
2005
2006

Cost of Materials
(percent of shipment value)

Cost of Payroll
for Production Workers
(percent of shipment value)

Cost of Total Payroll


(percent of shipment value)

52.78
12.66
52.67
11.94
53.89
12.38
44.67
8.89
46.11
8.48
U.S. Census Bureau. Annual Survey of Manufacturers, 2002-2006

12-5

28.51
27.26
28.32
22.12
21.92

12.3.1.2

Inventory Levels

Table 12.3 shows the year-end inventory levels obtained from the U.S. Census Bureau,
Annual Survey of Manufacturers, Statistics for Industry Groups and Industries from 2002 to
2006. Again, DOE presents these data as a broader measure of the BC and EPS industries. The
industries end-of-year inventory from 2002 to 2006 decreased when expressed both in dollars
and as a percentage of shipment value.
Table 12.3 All Other Miscellaneous Electrical Equipment and Component Manufacturing
Industry End-of-Year Inventory
End-of-Year Inventory
End-of-Year Inventory
($000s)
(percent of shipment value)
2002
1,847,565
26.75
2003
1,769,259
25.90
2004
1,738,681
25.08
2005
1,382,601
19.63
2006
1,376,288
18.98
U.S. Census Bureau. Annual Survey of Manufacturers, 2002-2006

Year

DOE obtained full production capacity utilization rates from the U.S. Census Bureau,
Current Industrial Reports, Survey of Plant Capacity from 2002 to 2006. Table 12.4 presents
production capacity utilization rates for NAICS code 335999. Full production capacity is defined
as the maximum level of production an establishment can attain under normal operating
conditions. In the Survey of Plant Capacity report, the full production capacity utilization rate is
a ratio of the actual level of operations to the full production level.
Table 12.4 All Other Miscellaneous Electrical Equipment and Component Manufacturing
Industry Full Production Capacity Utilization Rates
Year

All Other Miscellaneous Electrical Equipment and Component


Manufacturing (%)

2002
57
2003
62
2004
61
2005
70
2006
66
U.S. Census Bureau. 2007 Current Industrial Reports: Table 1a - Full Production Capacity
Utilization Rates by Industry: Fourth Quarters 2002 through 2006. November 2007

12.3.2 Interview Topics and Preliminary Findings - EPS


The following section summarizes information gathered during interviews held during
the fourth quarter of 2009 for the preliminary MIA.

12-6

12.3.2.1

Market Shares and Industry Consolidation

Amended energy conservation standards can alter the competitive dynamics of the
marketplace, prompting companies to enter the market, exit the market, or merge with other
companies. The preliminary MIA interview questions asked manufacturers to share their
perspectives on industry consolidation both in the absence of amended energy conservation
standards and assuming amended standards at various efficiency levels. The interview questions
focused on gathering information that assessed:

current and anticipated market share in the event of standards;


potential disproportionate cost increases to some manufacturers;
likelihood of industry consolidation;
increased proportion of fixed costs potentially increasing business risks; and
potential barriers to market entry (e.g., proprietary technologies).

The need to assess anti-competitive effects of proposed amended energy conservation


standards derives from the need to protect consumer interests. During the interviews, DOE also
solicited information to determine whether amended energy conservation standards could result
in disproportionate economic or performance penalties for particular consumer or user
subgroups. Manufacturers were also asked if amended energy conservation standards could
result in products that would be more or less desirable to consumers due to changes in product
functionality, utility, or other features.
Market Shares: DOE inquired about the current market shares of manufacturers in the
EPS industry and how those shares might change after amended energy conservation standards.
One manufacturer noted that new standards would not change market shares because improving
energy efficiency is now a primary design goal and most of the industry is already actively
marketing the efficiency aspect of their designs. A separate manufacturer believed that those
manufacturers that are already producing products close to the new standard levels could see a
short-term rise in market share until the rest of the market has time to adjust to the new
standards. One manufacturer expected some weaker companies to exit the industry due to the
new standards but did not expect that occurrence to have a long-term impact on market share.
Industry Consolidation: The EPS industry is relatively fragmented but has seen a
consolidation trend in recent years. Manufacturers were split on whether a new standard would
lead to any consolidation. Some believed that a standard could contribute a small amount to the
consolidation trend while others did not expect a new standard to have any effect on
consolidation. All agreed that any consolidation resulting from a new standard would be
insignificant.
12.3.2.2

Production and Product Mix

DOE requested manufacturers feedback on what they perceived to be the possible


impact of amended energy conservation standards on manufacturers products. New standards
could result in a change in manufacturers product mix or a change in the products utility to
consumers. Higher energy conservation standards could also result in higher per-unit costs that

12-7

could cause consumers to shift to less expensive alternative products, if such products were
available.
Product Differentiation: Manufacturers discussed the factors that differentiate the
products in the market. One manufacturer noted that product differentiators are changing because
of a maturation of the market which has been driven by increased communication. Prior to this
development, OEMs largely purchased EPSs from manufacturers with whom they had long
relationships. Due to the increased communication, trust in performance is increasing and OEMs
now base more of their purchasing decisions on price and brand.
Product Mix: The EPS manufacturers did not expect a new standard to have a
significant effect on the industrys product mix.
Product Utility: Manufacturers believed that new standards would have a positive effect
on product utility. One manufacturer stated that as EPSs become more energy efficient, they can
be produced in smaller sizes. This would improve consumers product utility. One manufacturer
disagreed, stating that new standards could result in larger enclosures for EPSs as well as less
flexible power cords. Both of these outcomes would decrease utility for consumers.
12.3.2.3

Conversion Costs

None of the manufacturers interviewed believed that a new standard would require
significant capital or product conversion costs. Manufacturers mentioned that they would expect
capital or product conversion costs to be very low for any of the proposed standards.
12.3.2.4

Cumulative Regulatory Burden

The mobile phone industry is in the process of attempting to implement a universal EPS
to be used with mobile phones. This initiative would allow consumers to purchase one EPS that
could be used with any new mobile phone. Because the mobile phone industry is currently
structured so that each mobile phone requires an EPS that is unique to the model, a universal
EPS has the potential to drastically reduce future shipments of EPSs for mobile phones. The new
universal EPS structure has a target date of 2012.
New safety regulations could force manufacturers of EPSs used in medical products
(designated as non-Class A EPSs) to provide more reporting on safety issues. One manufacturer
commented that the new safety standard could be implemented without much cost to
manufacturers.
12.3.2.5

Other Key Issues

The preliminary MIA gives DOE the opportunity to identify other key manufacturer
issues early in the development of amended energy conservation standards. During preliminary
interviews, manufacturers identified several major issues that are detailed in the following
sections.
Shipments: Shipment projections can be a significant factor in determining the
manufacturing impacts of amended energy conservation standards. The interviews provide an
12-8

opportunity for manufacturers to share information that can help DOE quantify the magnitude of
any changes in shipments resulting from amended energy conservation standards.
A few manufacturers stated that exceptionally stringent standards could cause costs (and
subsequently, prices) to reach a point where consumers could decide that the product was too
expensive. Potentially, this could cause consumers to migrate to lower priced products or delay
purchases altogether. However, none of the manufacturers expected the standards to be close to a
level where price would become a concern for consumers. All of the manufacturers felt that the
new standards would cause insignificant increases in product costs or even possibly lower
product costs. Neither scenario would have negative effects on industry shipments.
Timing: Some manufacturers expressed concerns that if new standards were
implemented too quickly, manufacturers may not have enough time to produce and sell all
remaining inventory at the prior efficiency level. In such a case, manufacturers would be forced
to write down inventory due to product obsolescence. One manufacturer estimated that the
company would need at least six months notice prior to a standard implementationsignificantly
less time than is usually provided for manufacturers to adjust to new standards. Another
manufacturer noted that the rulemaking needed to be relevant with the best technology currently
available. That is, because of the lag between the announcement of new standards and their
implementation, this manufacturer had concerns that new standards could be implemented after
the industry had already reached higher efficiency levels.
12.3.3 Interview Topics and Preliminary Findings - BC
The following section summarizes information gathered during interviews held during
the first quarter of 2010 for the preliminary MIA.
12.3.3.1

Market Shares and Industry Consolidation

Market Shares: DOE inquired about current market shares of manufacturers in the BCs
industry and how those shares might change after amended energy conservation standards. The
BCs industry is fragmented with few manufacturers owning more than a small portion of market
share. Within a few of the consumer product segments, a couple of manufacturers do control a
significant percentage of the market; however, as a percentage of the entire BC market, their
market share is much smaller.
Industry Consolidation: Due to the current state of the economy, many smaller
manufacturers are facing financial difficulties. One manufacturer commented that if the economy
continues to suffer, the smallest manufacturers may be forced to close. Multiple manufacturers
believed that new standards could adversely affect small manufacturers, especially if those
manufacturers are domestic.
12.3.3.2

Production and Product Mix

Product Differentiation: The vast majority of BCs are sold to consumers as part of a
larger consumer product. Because consumers make product decisions based on the product as a
whole, differences in BCs sometimes go unnoticed. One manufacturer stated that consumers

12-9

make purchasing decisions based on the product as whole while simply expecting the BC to
work well. As a result, the energy efficiency of BCs is not a primary consumer consideration.
According to multiple manufacturers, the main differentiator for consumers in selecting
products that utilize BCs is price, especially in a weak economy. One manufacturer stated that
product quality was the second most important differentiator. Although some manufacturers do
engineer their products to be more energy efficient, the manufacturers efforts are a result of their
own concerns about energy efficiency and are not driven by consumer demand.
Product Mix: The manufacturers of BCs did not indicate that a new standard would
have a significant effect on the industrys product mix.
Product Utility: Manufacturers of BCs believed that new standards would have a
positive effect on product utility, albeit minor. One manufacturer stated that as BCs become
more energy efficient, they can be produced to be smaller and lighter. This would produce a
positive change in product utility.
12.3.3.3

Conversion Costs

Several manufacturers stated that they expected to incur significant product conversion
costs to develop products that complied with new standards. The cost and time required for
safety approval and testing were mentioned by multiple manufacturers. If the standards were too
stringent, manufacturers commented that there could also be some capital conversion costs. One
manufacturer noted that if the standards were to push the market towards high frequency
technology, manufacturing could shift more to Asia, resulting in stranded domestic assets.
12.3.3.4

Cumulative Regulatory Burden

One manufacturer expressed concern that recent attempts by the US Department of


Transportation to harmonize safety and transportation regulations with those established by the
United Nations could require the manufacturer to add more circuitry to its BCs. Adding circuitry
would make it more difficult for the manufacturer to improve the energy efficiency of its BCs.
12.3.3.5

Other Key Issues

Uniformity and Clarity of Requirements: Almost all of the manufacturers expressed


concern over competitors finding ways to avoid the most stringent requirements of a new
standard. One manufacturer commented that at present, there are a substantial number of
competitors who do not attempt to get safety approval for their products and release them to the
market anyway. The manufacturers concern was that additional regulations without clear and
enforceable requirements could further disadvantage manufacturers who abide by the laws.
Another manufacturer stated that if testing requirements were too complex, some competitors
may find ways to avoid the most stringent requirements by exploiting loopholes or a lack of
clarity in the testing requirements.
Profit Margins: Multiple manufacturers stated that new standards could significantly
raise product costs and hurt their profit margins. A manufacturer explained that profit margins
were already exceptionally low due to the weakness in the economy. A different manufacturer
12-10

stated that the industry was unable to pass on increases in product costs to consumers.
Consequently, the industry faces decreasing profit margins whenever it experiences an increase
in product costs. Because cost is the biggest differentiator in BCs, if a manufacturer cannot
maintain its pricing structure, the manufacturer can rapidly lose market share.

12-11

CHAPTER 13. UTILITY IMPACT ANALYSIS

TABLE OF CONTENTS

13.1 INTRODUCTION ..................................................................................................................1

13.2 METHODOLOGY .................................................................................................................1

13-i

CHAPTER 13. UTILITY IMPACT ANALYSIS


13.1 INTRODUCTION
The U.S. Department of Energy (DOE will analyze specific effects of its proposed
standard levels on the electric industry as part of the notice of proposed rulemaking analyses,
using a variant of the U.S. DOE /Energy Information Administration (EIA)s National Energy
Modeling System (NEMS). The NEMS is a large, multi-sectoral, partial equilibrium model of
the U.S. energy sector. EIA uses NEMS to produce the Annual Energy Outlook (AEO)1. NEMS
produces a widely recognized baseline energy forecast for the United States, and this energy
forecast is available in the public domain. DOE will use a variant known as NEMS-BT to
provide key inputs to the analysis.a
The utility impact analysis will consist of a comparison between model results for the
base case and for policy cases in which proposed standards are in place. The use of NEMS-BT
for the utility analysis offers several advantages. As the official DOE energy forecasting model,
NEMS relies on a set of assumptions that are transparent and have received wide exposure and
commentary. NEMS-BT allows an estimate of the interactions between the various energy
supply and demand sectors and the economy as a whole. The utility impact analysis will report
the changes in installed capacity and generation, by fuel type, that result for each trial standard
level, as well as changes in electricity sales to the residential sector.
DOE will conduct the utility impact analysis as a policy deviation from the 2010 version
of the AEO (AEO2010), applying the same basic set of assumptions. For example, the operating
characteristics (e.g., energy conversion efficiency, emissions rates) of future electricity
generating plants are as specified in the AEO2010 reference case.
DOE also will explore deviation from some of the reference case assumptions to
represent alternative futures. Two alternative scenarios use the high and low economic growth
cases of AEO2010. The reference case corresponds to medium growth. As part of varying
supply-side growth determinants in these cases, AEO2010 also varies the forecasted energy
prices for all three economic growth cases. Different economic growth scenarios will affect the
rate of growth of energy demand.
13.2 METHODOLOGY
The electric utility impact analysis will consist of NEMS-BT forecasts for generation by
plant type, installed capacity, sales, and prices. NEMS provides reference case load shapes for
several end uses. The model uses predicted growth in demand for each end use to build up a
projection of the total electric system load growth for each region, which it uses in turn to predict
the necessary additions to capacity. DOE uses NEMS-BT to account for the implementation off
a

For more information on NEMS, please refer to the U.S. Department of Energy, Energy Information
Administration documentation. A useful summary is National Energy Modeling System: An Overview 2003,
DOE/EIA-0581 (2003), March 2003. EIA approves use of the name NEMS to describe only an official version of
the model without any modification to code or data. Because this analysis entails some minor code modifications
and the model is run under various policy scenarios that are variations on EIA assumptions, DOE refers to the model
by the name NEMS-BT (BT is DOEs Building Technologies Program, under whose aegis this work has been
performed). NEMS-BT was previously called NEMS-BRS.

13-1

energy conservation standards by decrementing the appropriate reference case load shape. DOE
will determine the size of the decrement using data for the per-unit energy savings developed in
the life-cycle cost analysis (chapter 8) and the forecast of shipments developed for the national
impact analysis (chapter 9).
The predicted reduction in electric capacity additions is sensitive to the peak load impacts
of the standard. DOE will investigate the need to adjust the hourly load profiles that include
battery chargers and external power supplies in NEMS-BT.
Since the AEO2010 version of NEMS forecasts only to the year 2030, DOE must
extrapolate results to 2055. DOE conducts an extrapolation to 2055 to be consistent with the
analysis period being used by DOE in the national impact analysis (NIA). It will not be feasible
to extend the forecast period of NEMS-BT for the purposes of this analysis, nor does EIA have
an approved method for extrapolation of many outputs beyond 2030. While it might seem
reasonable in general to make simple linear extrapolations of results, in practice this is not
advisable because outputs could be contradictory. For example, changes in the fuel mix implied
by extrapolations of those outputs could be inconsistent with the extrapolation of marginal
emissions factors. An analysis of various trends sufficiently detailed to guarantee consistency is
beyond the scope of this work and, in any case, would involve a great deal of uncertainty.
Therefore, for all extrapolations beyond 2030, DOE intends to use simple replications of year
2030 results; in this way results are guaranteed to be consistent. As with the AEO reference case
in general, the implicit assumption is that the regulatory environment does not deviate from the
current known situation during the extrapolation period. Only changes that have been announced
with date-certain introduction are included in NEMS-BT.
Results of the analysis will include changes in residential electricity sales and installed
capacity and generation by fuel type for each trial standard level, in five-year forecasted
increments extrapolated to 2055.

13-2

REFERENCES
1

Energy Information Administration, Annual Energy Outlook 2010 with Projections to


2035, March 2010. Washington, DC. Report No. DOE/EIA-0383(2010).

13-3

CHAPTER 14. EMPLOYMENT IMPACT ANALYSIS

TABLE OF CONTENTS

14.1 INTRODUCTION ............................................................................................................ 14-1

14.2 METHODOLOGY ........................................................................................................... 14-1

14-i

CHAPTER 14. EMPLOYMENT IMPACT ANALYSIS

14.1

INTRODUCTION

The U.S. Department of Energy (DOE)s employment impact analysis estimates national
job creation or job elimination resulting from possible new standards, due to reallocation of the
associated expenditures for purchasing and operating equipment. DOE will conduct this analysis
as one of the analyses for the notice off proposed rulemaking (NOPR). DOE will estimate
national impacts on major sectors of the U.S. economy, using publicly available data. DOE will
make all methods and documentation available for review.
Energy conservation standards can impact employment both directly and indirectly.
Direct employment impacts are changes in the number of employees at the plants that produce
the covered products, along with the affiliated distribution and service companies, resulting from
standards. DOE will evaluate direct employment impacts in its manufacturer impact analysis, as
described in chapter 12. This analysis covers indirect employment impacts which may result
from expenditures shifting between goods (the substitution effect) and changes in income and
overall expenditure levels (the income effect) that occur due to the implementation of standards.
DOE expects new energy conservation standards to decrease energy consumption, and
therefore to reduce expenditures for energy. The savings in energy expenditures may be spent on
new investment and other items. The standards may increase the purchase price of products,
including the retail price plus sales tax, and increase installation costs.
Using an input-output model of the U.S. economy, this analysis seeks to estimate the
year-to-year effect of these expenditure impacts on net economic output and employment. DOE
intends the employment impact analysis to quantify the indirect employment impacts of these
expenditure changes.
14.2

METHODOLOGY

To investigate the combined direct and indirect employment impacts, DOE will use the
Pacific Northwest National Laboratory (PNNL)s Impact of Sector Energy Technologies
(ImSET) model.1 PNNL developed ImSET, a spreadsheet model of the U.S. economy that
focuses on 188 sectors most relevant to industrial, commercial, and residential building energy
use, for DOEs Office of Energy Efficiency and Renewable Energy. ImSET is a special-purpose
version of the U.S. Benchmark National Input-Output (I-O) model, which has been designed to
estimate the national employment and income effects of energy saving technologies that are
deployed by DOEs Office of Energy Efficiency and Renewable Energy. In comparison with the
previous versions of the model used in earlier rulemakings, this version allows for more
complete and automated analysis of the essential features of energy efficiency investments in
buildings, industry, transportation, and the electric power sectors.
The ImSET software includes a computer-based I-O model with structural coefficients to
characterize economic flows among the 188 sectors. ImSETs national economic I-O structure is
based on the 1997 Benchmark U.S. table, specially aggregated to 188 sectors.2

14-1

DOE intends to use the ImSet model to estimate changes in employment, industry output,
and wage income in the overall U.S. economy resulting from changes in expenditures in the
various sectors of the economy. DOE designed the employment impact analysis to estimate the
year-to-year net national employment effect of these different expenditure flows.

14-2

REFERENCES

1. Roop, J. M., M. J. Scott, and R. W. Schultz. ImSET: Impact of Sector Energy Technologies.
2005. Pacific Northwest National Laboratory: Richland, WA. Report No. PNNL15273.
2. Lawson, A. M., K. S. Bersani, M. Fahim-Nader, and J. Guo. Benchmark Input-Output
Accounts of the U. S. Economy, 1997. In Survey of Current Business. December 2002: pp.
19117.

14-3

You might also like